Technical terms for 'House Elf Problem'

Question

There is an informal ethical dilemma known as the House Elf problem, after the Harry Potter slave race who are, universally*, enthusiastic servants who strongly prefer to be servile and be treated as lesser, and are happy to do dull, dirty, demanding work. House Elves enjoy existing, regard death the same way as humans do, and are generally happy.

The primary dilemma is "Is it ethical to create House Elves?" The secondary dilemma is "Given that they exist, is it ethical to make use of their services?" These are difficult questions because once they exist, they strongly prefer to continue existing and strongly prefer to have their services used, and before they exist they have no preferences, and yet many people's moral intuitions oppose making use of slaves even if they're happy slaves.

Obviously the broad strokes of "consider that a natural slave race exists" goes back to Plato, but is there a more formal/academic terminology for the modern variant, where we're considering the possibility of creating a slave race? I couldn't find anything in the SEP, or searching for papers.

*For the purpose of the philosophical problem, it's normal to posit that outliers, like HP's Dobby, don't/won't exist.

Answer 1

Along with the sorcerers apprentice and spells that work too well, I'd interpret the House Elf Problem in terms of AI. Specifically, Bostrom's Malignant Failure Modes, and in the House Elf case look at whether it could be considered 'mindcrime'.

Defenders of House Elfs as slaves saying they really enjoy it, should consider the Sonmi chapters of David Mitchell's 'Cloud Atlas'. "Nymph in thy orisons be all my sins remembered." Replicants in Blade Runner with implanted memories are another case.

Various philosophers have sought to justify slavery, including Aristotle, Aquinas, and Hobbes. Defenders of slavery in the era when it was being banned often claimed slaves were better off in that condition. The beginning of chattel slavery was war enslavement as an alternative to death. Genocide of the defeated is not defended anymore, though.

Slavery is generally considered the main reason Rome didn't have an industrial revolution despite having many precursors for one - it's been worked out the average living standards of a citizen in Rome were not matched again until London in the 1720s. US cotton production increased after slavery ended.

There's also a game-theory argument, about differential selection, discussed in relation to tyrannicide here: Is the tyrannicide perpetrated by William Tell morally legitimate?

Another angle is postcolonial theory. This answer discusses epistemic justice, the potential distortions in subjugated people's perceptions of reality and truth-making: Need help with this paper on epistemic justice And, this is a comic take on applying postmodern analysis to whether races have essential or intrinsic qualities, but is philosophically accurate, and relates to a context of fantasy literature: Existential Comics on whether orcs are evil, or elves are racists.

DeBeauvoir particularly clearly covers oppression of people complicit in their oppression, in The Second Sex. Discussion here: Evil Force in The Second Sex mentioned by Simone de Beauvoir

I'd say the deepest philosophical thinking on different types of being, is from Buddhism. For Buddhists all minds have an intrinsic quality of capacity for awakening, and should be treated with compassion and supported wherever possible to develop spiritually, as the only aspect of being with lasting consequences.

Indra's Net is a metaphor used in Buddhist thought to illustrate dependent origination, that can help us understand how non-grasping awareness in interaction leads to mutually-arising intersubjective reality. Discussed in more detail here: According to the major theories of concepts, where do meanings come from?

This perspective can also be linked to Peter Singer's framing of moral progress as expanding the circle of moral concern, that we have moral obligations to respect capacities of other creatures rather than essentialise only humans as moral subjects, and so also miral obligations to support the increase of capacities (intellectual, miral etc) of other species (this is comoarablevto the bodhisattva path of working towards awakening of all beings).

So in summary, sharing the intersubjective space with other minds, expands the reality of all the participants through their sharing of experiences. We can understand this increasing participation with others as the direction of moral progress (I'd also look to Maslow's self-actualisation, and development of what we can be through interactive play If you had all you want, would you be eventually happy or bored? which requires participant's basic needs to be secure, & a non-exploitative interaction). The alternatives to interacting as equals, or in a capacities/responsibilities framework, involve the potential immorality of mindcrimes, and threats related to malignant failure modes of AI (which can also be related to game-theory & the social contract). We can also relate the intrinsic dignity & scope of minds to DeBouvoir's understanding of authenticity, & Buddhist capacity for awakening.

Answer 2

The "house elf problem" does not, to my knowledge, yet have a name. It is a good distillation of a general problem with individualistic/rights-based ethics.

The general issue is based on the reality that individuals are not the only, or often even the best, referent that one should use for either moral or other optimizing framing. If, for instance, one is interested in species preservation, then the welfare of a single post-breeding individual of the species should not be a concern. If one is interested in the health of Gaia, the biosphere, then even the continuation of an entire species would be of minimal concern, because Gaia operates with regular species replacement.

For we humans, the welfare of a household, an extended family, or a small community is often of far more importance than the welfare of each individual member of a community. Contemporary western moral thinkers operate within a massively successful community, where there is significant survival margin and therefore a high tolerance margin to address secondary concerns such as freedoms and self-actualization. If we lived as most humans have in history -- one the edge of starvation, then many of the intrinsic prejudices we humans have which are problematic for our more modern and comfort-based moral sensibilities, would make much more sense.

For instance, we humans are somewhat eusocial -- we support the others within our community in times of hardship. This is a moral virtue, and increases our ability to individually survive erratic fates, but it imposes a carrying cost on the community for everyone who is seen as "us". Strangers, foreigners, whose commitment to supporting "us" themselves when the rest have hardship is unclear, SHOULD be regarded with suspicion when resources are scarce. Even more brutal, those who are disabled from birth, "cripples", were a burden that communities often could NOT afford to support for a lifetime -- and infanticide through exposure was often a justified community response to reduce the risk of wholesale starvation due to privation.

In this context, that social advantage sometimes trumps individual disadvantage, we can discuss the consequences of the reality that humans generally have to organize hierarchically to do things done that require more than a very few of us to get done. One of the consequences of humans having innate tendencies to organize hierarchically, is that we on occasion SEEK strong leaders, rather than always pursuing individual self-actualization. Followership is built into us as part of our evolutionary makeup. All of us have some degree of house-elf in us.

We can see this more clearly in creatures that are more communal than people are. Horses, for instance, are happiest when they are in a herd, with a strong domineering herd leader. This is something that horse trainers can take advantage of -- making THEMSELVES the herd leader for the horse, assists tremendously in training a horse for useful/safe riding. Dog training uses similar principles, a well-trained dog recognizes their people are the pack leaders.

House elves take these principles to a logical extreme -- the HOUSEHOLD becomes the unit of value, the humans are the heads of the hierarchy, and the house elves ARE happiest when purely subservient. This is different from dogs and humans. For us and dogs, each of us may at some point need to serve as leader, hence we each have both leadership and followership traits. Plus dogs and humans also operate outside the household/pack, so must be able to operate competently autonomously/equally as well as hierarchically/communally. Making subservience a species trait, not just a temporary role -- well we humans did that to wolves to make dogs...

At any rate, hierarchies, subservience, followership, and self-effacement are crucial traits for communities and organizations to thrive. They are "problems" for purely individualistic ethics. But if one can change one's framing, adapting one's reference system to different scales: from genes, to cells, to individuals, to communities, to species, to cultures, to ecosystems, or to Gaia, as appropriate -- then followership and "menial" roles are no longer a problem, but a benefit.