I have long been intrigued by Orson Scott Card’s typology of relations to the other, as expressed in his novel Speaker for the Dead. I like it so much that I used it as a central part of my argument (in Chapter 2 of this forthcoming edited volume) that the depicted relations between Colonials and cylons in Battlestar Galactica can tell us something about how to construct a more humane social order. But teaching the novel for the umpteenth time in my sci-fi course this week — it and its prequel Ender’s Game have been in every iteration of the course since I wrote the first draft of the syllabus when I was about fifteen — it occurred to me that the Card typology needed some analytical tightening before it could be truly useful. I did a first run at working through the issues in class on Wednesday; this post is my second attempt. And it has prettier diagrams than the ones I scrawled on the whiteboard in class.

Card’s typology contains four orders of otherness or foreignness:

The first is the otherlander, or utlanning, the stranger that we recognize as being a human of our world, but of another city or country. The second is the framling…This is the stranger that we recognize as human, but of another world. The third is the raman, the stranger that we recognize as human, but of another species. The fourth is the true alien, the varelse, which includes all the animals, for with them no conversation is possible. They live, but we cannot guess what purposes or causes make them act. They might be intelligent, they might be self-aware, but we cannot know it.

The major advantage of this typology, and the potential that I find in it, stems from the fact that it is a typology of relations, not a typology of already-fully-formed communities.
There is no danger here of reifying a contingent social arrangement, and imbuing it with more stability and solidity than it warrants; a community is, in this approach, more or less completely endogenized to the relations between and among its members. Community is a function of how we relate to one another, and is as such located “between” us rather than “outside of” us. The other thing I particularly like about this typology is that it does not make the self a residual effect of differentiation from the Other; exclusion and separation are not the fundamental or foundational aspect of identities and communities, but emerge instead as one possible mode of relation. There can thus be identities and communities that are not premised on deliberate exclusion as much as they involve a focus on “internal” matters — a potential that often gets minimized or sidelined in certain kinds of poststructural and critical accounts of identity. (There’s a very nice 2005 APSR article by Abizadeh — couldn’t find a link that wasn’t behind a paywall, sorry — that makes this point quite cleanly.)

That said, there are three problems with the Card typology. First, although he refers to it as a “hierarchy of exclusion” and thus implies a single criterion that gets more or less intense as we move up and down the scale, the typology actually contains two quite different logics and thus envisions two very different kinds of communities emerging from two different kinds of relations. Second, and related, Card hasn’t actually presented four different types of relation; he’s only presented three (utlanning and framling actually collapse into one another), and only two of them are related to possible communities (no conversation with the varelse means no community with the varelse). Finally, because Card treats this as a classification of actually-existing relations rather than as an ideal-typical typology of possible relations, he is unable to really probe the intriguing dilemmas that arise in actual communities that are never completely characterized by any one kind of relation. The right question is not “what kind of relations to the other do we have in our community?” but instead is something like “what kind of community is envisioned by our ways of relating to the other?” and the reformulated ideal-typical typology helps us sort out the answer to that second question.

I’ll take these in sequence. First of all, notice that both utlanning and framling are relations that presume a shared connection between the people involved, and that the shared connection in question is more or less “backgrounded” or naturalized in the course of any interaction that people engaged in performing or enacting that relation have. In Card’s terms, this prior connection is that both parties are “human,” but I want to suggest that the specific character of the prior connection is less relevant than the fact of its existence as a taken-for-granted presupposition. Relations of the utlanning and framling type actually both look basically the same:

In this diagram, A-D are the relating parties, and for the sake of visual clarity I have drawn their network of relations as a maximally connected one. The green circle around A-D represents the common membership of all of the relating parties in some broader set characterized by something that they all have in common — some attribute or feature by virtue of which they are all members of a single category that is larger, conceptually speaking, than each of them themselves. Call this a categorical community: differences between the relating parties are in a sense overshadowed by their common membership in a broader community, and note that in order for this to work the category in question has to be socially/culturally/politically salient — mere physical resemblance will not cut it. In fact, in a categorical community, the relating parties in some sense think of themselves as belonging to that broader category; if they do not, it matters not at all whether the relating parties share some characteristic in common. (I am setting aside for the moment the complexity of how one recognizes common characteristics and how they become socially/culturally/politically salient, because I am ideal-typifying the kind of community envisioned by the utlanning and framling modes of relating to the other. My point is that in utlanning and framling relations, there in some sense already is a salient category.)

The reason that this diagram could be both utlanning and framling — and, if we get less interplanetary about things, it could be states within a region or civilization, towns within a nation, individuals within an organization, etc. — is that utlanning and framling relations only differ from one another in their physical and geographic size. Utlanning relations are bounded by a planet, which would be the green line in the diagram; scale up, make A-D planets instead of states or cities or countries, and for have framling relations. Scale down to get one of the other varieties I mentioned a moment ago. All of these are relationally similar, and all feature the same basic arrangement: the relating parties may differ from one another on a whole variety of issues and attributes, but their commonality provides a broader basis on which to recognize one another as in some sense belonging together.

Belonging to a categorical community makes possible one of the most striking (and arguably one of the most effective) types of rhetorical coercion: the “nesting” gesture in which a speaker appeals to the broader category as part of a socially sustainable argument about a proposed course of action. In some of my work I have explored this gesture in some detail, as nesting is what makes appeals to “the West” so politically effective during the post-WWII era and well into the Cold War. Nesting is the inner logic of claims about civilizational identity, regional solidarity, national unity, etc. In all of these cases and many more besides, nesting works as a rhetorical gesture because the relating parties in some sense understand themselves to belong to a larger category, which makes possible the appeals to that category that can function as a rhetorical trump card in a public debate: we have our differences, but now the tribe/nation/planet/species to which we all belong has need of us, so we have to put aside those differences for the sake of the greater whole.

Lose the salient category and you lose this possibility, along with the categorical community itself. What you have instead is a category of relating parties that only share in common their ability to relate to one another, and nothing more fundamental than that:

Again, for visual clarity I have drawn this as a maximally connected network, but it need not be. The key point is that in this kind of noncategorical community there is no agreed-upon categorical boundary within which and on the basis of which all the parties can relate; there is just the web of relations itself. Obviously there’s no nesting here.

This is, I would say, a diagram of the community envisioned by Card’s raman relations. The word “human” in “human of another species” doesn’t mean the same thing as it meant in Card’s definitions of utlanning and framling relations; indeed, “human” in raman relations seems to mean nothing other than the ability to converse and communicate, since the primary conceptual distinction is between raman and varelse on the grounds that varelse relations do not include the possibility of conversation. (“They live, but we cannot guess what purposes or causes make them act. They might be intelligent, they might be self-aware, but we cannot know it.”) There would thus be no community envisioned by varelse relations, and no categorical community envisioned by raman relations.

Of course, any student of world politics will have already recognized the parallels here: a categorical community is like “domestic politics,” and a noncategorical community is like “international politics.” Unlike Waltz, what I am prioritizing here is categorical membership, not hierarchy/anarchy, but the basic opposition is the same: presumptive commonality inside the state, and something much more explicitly process-dependent outside. (Again, remember that I am not describing actually-existing entities yet. Actual communities inhabit the tension between these ideal-types.) Precisely because there is no readily-available salient common category to appeal to in a noncategorical community, relations have to be conducted on the basis of ongoing diplomatic negotiation, and the only controlling authority would be the previously-agreed-upon consent of the contracting parties, a.k.a. “precedent.” (This is perhaps starting to sound even more familiar to IR folks.) And any hierarchy would have to be either illegitimate or a-legitimate — or a temporary exigency that was extremely fragile, dependent on independent calculations of benefit and interest…

So what have we learned? Raman relations, which envision a noncategorical community, are similar to what we IR scholars often imagine to be “international relations” — states in anarchy, sure, but more to the point, independent entities that only share in common the ability to relate to one another. (I am quite deliberately shifting to the more generic “relate” rather than Card’s “conversation,” because there is an implicit teleology in Card’s formulation: “conversation” makes violence less likely, because — and this is a general theory of Card’s — understanding makes conflict evaporate. I am not so sure about this, and more to the point, my argument doesn’t depend on it: the “diplomacy” in a noncategorical community could easily consist of an exchange of gunfire, as long as the various parties involved thought of one another as actors rather than as natural forces or something.) Utlanning and framling relations, which envision a categorical community, are similar to what we IR scholars often imagine to be “domestic politics,” with the hierarchy that we sometimes think of as characteristic of domestic politics linked to legitimation claims involving the rhetorical gesture of nesting. And encounters with “true aliens,” the varelse relations: no community is envisioned. Three categories, not four, and two logics: a logic of scale for utlanning-framling, and a logic of recognition for raman-varelse. (And note that Card implicitly realizes this; at the beginning of the first chapter of Speaker he has the author of the typology point out that “When we declare an alien species to be raman, it does not mean that they have passed a threshold of moral maturity. It means that we have.”)

I presume that anyone who has made it this far into this essay has already raised about a hundred objections to the distinctions I am drawing here, all of which boil down to some form of “but actual communities don’t look like this!” Precisely. I agree. The point of an ideal-typical typology is not to describe actually-existing objects or entities or relations; the point is to tease out certain logical relations which can then be used instrumentally in the course of an explanation of some puzzling actual situation or case. It is doubtful, for instance, whether there is any such thing as an actual noncategorical community; international society developed a “standard of civilization” detailing who could play and who could not play, and liberal societies with roots in a social-contractarian understanding of social order and government have tended to rest, even if only implicitly, on some categorical community to determine whose voices needed to be taken into account: white people, men, property owners, etc. Indeed, I would speculate that, precisely because categories are so useful in managing daily social relations, a noncategorical community, by virtue of repeated transactions among its members, might incline toward being a categorical community at some point, as raman relations gave way to utlanning or framling relations. But by the same token, a categorical community might break down if one of the relating parties did something that was simply so beyond the pale that it appeared to be the action of an inhuman alien (utlanning/framling -> varelse) or, less dramatically, something happened that revealed that the presumptively stable category on which the community was based was in fact considerably more evanescent (utlanning/framling -> raman). How particular people react to changing situations, and how particular institutions and practices function, would then be explicable in terms of the tension between these different visions of community and the different relations that give rise to them.

The reason this isn’t just redescription, or recoding in exotic language, is that the perspective I have sketched here does three things with respect to thinking about identity and community that other typologies do not do, or at least do not do as well:

1) community here is a function of social relations, and not an ex ante presumption. The kind of community that is envisioned depends on the kind of relations its members have to one another.

2) community here is not necessarily premised on exclusion — what matters is less who we don’t incorporate, and more how we relate to one another. A categorical community can exclude, sure, but that exclusion is a consequence, not a cause.

3) the important empirical questions to ask about a community are not where its boundaries are and how it maintains them. The important empirical questions are instead: a) how is the category on which a categorical community is based made to seem natural and inevitable to the members of that community? b) how is the determination made that a given other is raman and not varelse, or vice versa? (In this last register especially, I wonder about the principle of indefinite detention and the preference for “surgical” drone strikes and other targeted killings of alleged terrorist masterminds, but I have gone on long enough so I will leave that as an exercise for the reader.)

Final thought. The idea of a noncategorical community, although it may lack an empirical referent, serves an important moral function. If we are actually committed to the idea of a community in which the members share in common only their capacity to converse and communicate — their moral worth and practical agency — then we have to be perpetually on guard against the all-too-human tendency to fill the “human” in “human of another species” with specific categorical content. It matters not at all whether that content comes from some presumptively natural attribute (as though the question of what constitutes a human or a sentient being with moral worth could be definitively answered in some objective manner) or from some presumptively transcendental exercise of pure communicative reason (Kant and Habermas, you’re on notice here) or doctrinal revelation (the traditional “faith-based” solution to this conundrum is to have some deity show up and simply tell you who is worthy of moral recognition, and perhaps more importantly, who is not worthy). Instead, if we have the courage of our convictions, we need to oppose any and all attempts to unequivocally bound the human, rather than staking everything on some categorical definition of who and what we are. Perpetual negotiation in a noncategorical community of raman relations may be uncomfortable and murky, but such a vision — one might even call it a prophetic provocation — might be the only way to preserve our humanity, and to stay open to possibilities yet unknown.