In section 11.2, Boyer acknowledges that all languages have “verbal forms” that express ownership in one way or another, but he also claims that “it would seem difficult to infer conceptual structures concerning ownership from these forms,” and that the meanings of such forms are “not sufficient for representations of possession, ownership, and property.” In contrast, I maintain that cross-linguistic patterns involving ownership provide a unique source of insight into the psychology of ownership, with special relevance to selfhood.

As Boyer notes, English uses the same suffix to indicate possession of all kinds of entities, including body parts (e.g., John’s foot), blood kin (e.g., John’s father), affinal kin (e.g., John’s wife), attributes (e.g., John’s pride), associations (e.g., John’s country), and objects (e.g., John’s car). He also mentions that many other languages use separate grammatical constructions to distinguish between two conceptually distinct types of possession that are usually called “alienable” and “inalienable.” Unfortunately, however, he neglects to consider the most interesting aspect of this distinction – namely, that even though different languages draw it in different ways, such variability is far from random, since the inalienable category tends to be reserved for possessions that are construed as integral to their owner's identity, whereas the alienable category covers all sorts of unessential possessions.

For present purposes, the most pertinent findings are as follows (for detailed data and discussion, see Aikhenvald & Dixon, Reference Aikhenvald and Dixon2013; Chappell & McGregor, Reference Chappell and McGregor1995; Dixon, Reference Dixon2010; Nichols, Reference Nichols and Shipley1988; Stolz, Kettler, Stroh, & Urdze, Reference Stolz, Kettler, Stroh and Urdze2008; Velazquez-Castillo, Reference Velazquez-Castillo1996). First, body parts are almost always marked as inalienable, with exceptions typically limited to detached body parts, bodily excretions, and body parts that are uncontrollable, like hair. Second, some languages extend the domain of inalienable possession to blood kin, and some extend it further to affinal kin. Third, some languages incorporate into their inalienable category certain attributes of a person, such as physical properties (e.g., height, smell, shadow), mental states/traits (e.g., pride, honesty, anger), or other characteristics (e.g., fame, wealth, luck). Finally, some languages also treat as inalienable particularly important associations of a person, such as certain non-kin social relations (e.g., friend, master, shaman), certain habitats (e.g., house, homeland, grave), or certain artifacts (e.g., canoe, spear, ring).

Taken together, these cross-linguistic findings suggest that inalienability reflects what Bally (Reference Bally, Frankhauser and Jud1926/1995) calls the “personal sphere,” a malleable realm that embraces, for a given speech community, possessions conventionally construed as being central to selfhood. More precisely, the data point to a complex self system with two components. First, the core self is grounded in body ownership, and this may be why the kinds of possessions most frequently marked as inalienable are body parts, with some languages even restricting their inalienable construction to these objects. Second, the extended self includes noncorporeal possessions that are vital to one's identity, and this may be why many languages apply their inalienable construction not only to body parts, but also to other classes of personally significant entities, like family, friends, and certain kinds of valued things. This notion of a complex self system fits nicely with William James’ (Reference James1890, p. 291) observation that “between what a man calls me and what he simply calls mine the line is difficult to draw. We feel and act about certain things that are ours very much as we feel and act about ourselves” (see also Belk, Reference Belk1988, Reference Belk1991). It's also notable that the neural substrates of the core and extended components of the self have recently begun to be studied (Abraham, Reference Abraham2013; Aglioti, Smania, Manfredi, & Berlucchi, Reference Aglioti, Smania, Manfredi and Berlucchi1996; Blanke, Reference Blanke2012; Ehrsson, Reference Ehrsson and Stein2012; Kim & Johnson, Reference Kim and Johnson2014; Serino et al., Reference Serino, Alsmith, Constantini, Mandrigin, Tajadura-Jimenez and Lopez2013).

These points have implications for Boyer's model. For one thing, while his “P-tags” do include a parameter for the strength of the possessive relationship, the model as a whole does not explicitly recognize that some owned things are typically treated as more significant than others. As described above, the cross-linguistic patterns suggest – in keeping with other findings from psychology and neuroscience – that the most precious owned things tend to fall within a complex self system that includes not only the corporeal self but also the extended self, which encompasses certain closely related people and objects. In addition, even though the cross-linguistic patterns don't directly reflect any of the intuitions about ownership that Boyer seeks to explain, speakers have clear intuitions about how alienable/inalienable possessive constructions should be used, and these intuitions are based on social conventions or norms regarding the conceptualization of ownership for communicative purposes. Finally, one goal of Boyer's model is to explicate the “general respect” for ownership that community members take for granted. In this context, the cross-linguistic data lead to an interesting question: In communities where the dominant language requires that certain treasured objects are obligatorily marked as being inalienably possessed by their owners, do people accord those objects greater respect than objects that are inalienably possessed? I'm not aware of any research on this topic, but it seems relevant to Boyer's concerns.