In my previous post, I presented an account of moral status in terms of final value and direct obligations. About a year ago, another featured author, David Killoren, presented his "Extended Narrativity Hypothesis" (ENH) to acount for a the different obligations we have towards livestock and our pets (which he sums up elsewhere as a reformed "animal caste system") (see more details here and here). He wrote,
If the Extended Narrativity Hypothesis is true, then there are at least three categories of morally important beings, which I label as follows: self-actualized persons (including intellectually normal human beings) who possess non-fungible value in virtue of their self-narration of their own lives; persons-by-proxy (including pets) who possess non-fungible value in virtue of others' narration of their lives; and non-persons (including livestock) who do not possess any non-fungible value.
This three-way division of morally important beings can be used to explain why humans, livestock, and pets seem to belong to distinct moral categories. This is a big advantage for the Extended Narrativity Hypothesis, in my opinion.
In the wake of my previous post, I'd like to offer a reply to David. While I am very sympathetic to the attempt to account for different obligations based on other features than just an animal's intrinsic properties, and while the ENH seems to plausibly account for some widespread intuitions about pets and livestock, I take issue with the way David frames the kind of value grounding our obligations and with his argument for generally excluding livetsock from the realm of narrative value.
David's two-tier system only provides an account of different obligations towards animals (such that more stringent constraints apply to how we treat our to pets), in terms of what one ought to do period, rather than an account of our obligations to them, in terms of what one owes them. I believe David would agree. Yet I don't believe this understanding of our obligations fully accounts for the intuitions he takes his hypothesis to account for. If my account of moral status is correct, and if the kind of caste David describes captures something like moral status, then we owe our pets more than just responding appropriately to a generic kind of value. We owe particular individuals specific treatments — our obligations are directed to valuable individuals. Merely responding to a type of value without regard for a particular token individual signals a failure to understand what that value implies: attending to an individual for its own sake. (This is why I take moral status to imply final value in concrete objects rather than Moorean intrinsic value supervening on states of affairs.) Of course, the particular relations that such obligations reflect are, coincidentally, made salient by a narrative outlook. Framing our relationships not only as stories about those we care about, but also as overlapping narratives, makes it salient that our special obligations are de re obligations connecting particulars (say, you and your dog), rather than de dicto obligations about whoever happens to instantiate non-fungible value (this is not to say that de dicto obligations to whoever fills the place of, say, your future companion, friend, or child are not important). But the story is not the whole story about our obligations.
David's attempt to establish the reformed animal caste system he envisions fails because there is no principled reason to think of livestock and pets as distinct natural kinds enabling different kinds of responses and valuations. Whatever they really are, such categories are plastic and context-dependent (think of how dogs are entitled to wildly variable treatments across cultures). The only way we can keep using these categories in morally relevant ways, that is, as picking out morally relevant features, will likely exclude whatever differences make it the case that the ENH applies to pets but not to livestock. If, ultimately, members of either category could end up earning (or losing) narrative value by proxy, why keep tying different kinds of value to different categories rather than just ask under what conditions the ENH applies? Since there is little reason to think the ENH cannot and should not apply to animals as a matter of fact referred to as livestock, the ENH offers an explanation of the non-fungibility of any animal whose story we can tell rather than an account of a reformed animal caste system.
But the ENH proponent has to bite is actually sweet. While we don't have decisive reasons to promote more narrative value (as David plausibly claims), still the constraints that (by-proxy) narrative value gives rise to highlight the significant loss that failing to recognize it could lead to (i.e. if we treat our companions as fungible). If the possibility of such loss matters gravely, and I think it does, why should we ignore the fact that many people actually (or could potentially) care, altruistically, about how well the singular life of any particular farm animal goes? The explanation of livestock's lacking by-proxy narrative value is that we have brought them to existence in order to eat them and/or their products. According to the ENH, having morally special relationships with them (of the sort that allows for narrativity) would twart such a purpose by making them non-fungible. In fact, it seems plausible that we (as a society, as corporations, farmers, or consumers) have a vested interest in preventing non-fungible value from emerging. And it could be that, under a reformed caste system, this would be optimific. But I see several problems with this:
(A) How could creating animals with a view to preventing them from earning non-fungible value be a practice worth preserving? It the anwser is: because this would achieve optimific outcomes, why not just do away with narrative value? If the reason against creating more narrative value is that it fails to maximize fungible value, why not prevent narrative value from arousing in future pets (i.e. abolishing animal companionship)? Granted, the fact that narrative value is worth conserving (when it exists) does not entail that it's worth creating or promoting (sometimes quite the contrary). Yet if one doesn't want to claim that there's something good about there being narrative value at all, the caste system, and the status of pets, may just as well gradually disappear (on pains of becoming an ad hoc justification of biases and prejudices).
(B) It's plausible that the only way to make the reformed system acceptable is by treating livestock in a way that ipso facto allows them to have by-proxy narrative value, i.e. by not treating them as mere resources, recognizing their individual needs and personalities, providing individualized care, letting them live longer, happier lives, etc. This is what so-called humane farming purports to be about, and such farming purports to be the only viable alternative to factory farming (which is wrong on David's view). Now, the value of all these actions can hardly be reduced to pure hedonic value. Or, to the least, these things bear the seeds of narrative value in addition to hedonic value. Therefore, either all sentient animals have both hedonic and narrative value, or none does.
Our alternatives are as follows:
- Narrative and hedonic value are both fungible, relatively to one another and to some extent only, e.g., in cases in which not all narrative value can be preserved, or when narrative value is dwarfed by hedonic value;
- Both narrative value and hedonic value are absolutely fungible;
- Neither narrative value nor hedonic value are fungible (e.g., because the latter is indexed to particular individuals that cannot be replaced without loss).
David obviously rejects (2), and for his take on narrative value to be fully compatible with the fungibility of hedonic value and the importance of optimific outcomes, I think he is committed to (1). (3) sounds at least as appealing as David's view insofar as (3) accounts both for the intuition that pets have non-fungible value and, for instance, the moral quality of rescuing and sending particular farm animals to sanctuaries. Is there a distinctive feature about rescued animals that makes them irreplaceable while their counterparts are not? The reason why they were rescued in the first place is because of their value, rather than a new source of value. And such reasons would, all things being equal, apply equally to those who could not be rescued: even more fundamentally, sanctuaries may be seen as a transition towards a more radical reform of our relationships to domesticated animals rather than just a way to make amends to a few lucky ones (see e.g., here and here). Farm animal rescues, at a small scale, in fact capture the potential scope of extended narrativity (the tension between the individualized life-stories of animals raised for food and the permissibility of killing them is explored — in the case of two oxen, Bill and Lou, at Green Mountain College — by James McWilliams in the Politics of the Pasture.). It's because these animals each have their own particular life stories (if only waiting to be realized by proxy) that they should not be seen and treated as fungible. It's not (only or primarily) because they happened to be rescued that they earned that kind of value.
Finally, the reason why the ENH applies to pets even though they're not persons lies in the possibility to meaningfully tell stories about their lives. Similar stories could not be told about chairs, cars, trees, and oysters, at least not the kind of stories that expresses their non-fungibility in terms that are not strictly agent-relative or metaphorical. (Narrative considerations can certainly apply to such cases, but I still think David would agree that a pet's value and a tree's or car's value in virtue of their life stories are of different kind.) Now, the reason why it's distinctively meaningful to tell a dog's story is because the dog, while not being a person, is conscious and arguably has a relatively temporally extended form of self-awareness: although the dog does not consciously construct (let alone narrate) a his or her life-story, the dog's life is not a mere fiction — it results from the dog's conscious desires, experiences, and actions. Similarly, most if not all current farm animals have behavioral and cognitive characteristics sufficient for having their own life stories. Cars and trees don't.
Therefore, why we less readily attribute narrative value to the lives of farm animals is just because of the particular socio-economic factors that prevent us from seeing them as individuals. The animal caste system, even a reformed one, appears to simply reasserts this arbitrary separation. Recent work in psychology actually suggests that motivated cognition leads us to deny minds to animals we (are led to) consider as food as a way to reduce cognitive dissonance. We strongly believe that it is wrong to cause unnecessary suffering. We could refrain from eating animals we see as minded (where having a mind implies being sentient), or we could eat them and not see them as minded. But eating animals we know can suffer while believing it is wrong to cause unnecessary suffering (tacitly assuming eating animals implies causing animals to suffer unnecessarily) is uncomfortable. It's plausible that a similar explanation could debunk the intuitions regarding the respective values of pets and livestock.
To conclude, the ENH explains why we have strong obligations towards our pets but not why such obligations could not apply to all farm animals (not just the particular ones we happen to bond with). Additional obligations bind us to our pets, given their specific needs, abilities, and the context in which they entered our lives, but they do not stem from narrative value as such. Narrative value gives rise to strong obligations towards all animals except for those who could not possibly become companions or earn narrative value by proxy. Perhaps some fully wild animals — if any, and supposing there is such a thing — could fit into this category. But perhaps our intuition in this case is as unreliable as our intuitions in the case of pets and livestock.
A further consideration in favor of grouping all domesticated animals together boils down to one explanation rejected by David, namely, that special obligations arise from relationships. All domesticated animals bear some relationships to us that most non-domesticated animals do not: relationships of (inter)dependence, vulnerability, mutual trust, shared or overlapping communities, and so on. Not surprisingly, narrative value maps nicely onto those relationships, although I would not go as far as to claim that the latter exhaust the potential for narrative value. Either way, relationships or narrative value apply to both livestock and pets. The ENH provides new insights into the ethics of our complex relationships to different animals that are otherwise intrinsically similar, but the story is to be continued.