Abstract
Many trans women (men) say that they know that they are women (men). Anti-trans activists deny the claims trans people say they know. Many say that social kinds like woman, Latinx, and consent are in some important sense constructed in the social world and are thus open to a certain amount of engineering. I think the claims to knowledge trans people make are correct, and I think it correct that such things as gender, race, and consent are constructed by society and so are prime candidates for what philosophers these days call conceptual engineering. But it is not all that easy to see how the claim about knowledge and the claim that what is known is determined by the vagaries of the social world are to be reconciled. In this paper I argue that this is a real problem, that it has a solution, and that the solution tells us something important about what happens when we contest norms or engineer concepts.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
A more rigorous statement of this argument would need to take a stand on whether ‘Mika is a W’ expressed the same proposition at different times or (since it says, roughly speaking, at a that Mika is now a W), or whether it expresses different propositions at different times. The argument goes through either way.
The issue here is reflected in Cappelen’s characterization of the consequences of his view. Suppose t is a time before the engineering of gender words and concepts and Mika utters ‘I am a woman’ at t. Cappelen holds (as a result of his views on the semantics of attitude ascription) that we speak accurately after engineering if we say that when Mika spoke at t, our protagonist said that (she knew that) she was at t a woman. But he also holds that we must say that when she spoke, she spoke falsely. Cappelen says that this aspect of his view is ‘a feature, not a bug’ (Cappelen 2018, p. 114). Somehow, I don’t think trans people will see it this way.
See (Jackman 1999) for a development of the view that future turns in usage can have an effect on current semantics.
I draw no distinction between meanings and concepts here.
This is, by the way, perfectly consistent with the idea that there are multiple more narrowly defined kinds. Reform Jews recognize multiple kinds of Jews. The Orthodox (should) recognize that there are multiple kinds whose membership is identical with that of the kinds which the Reform label ‘Orthodox Jews’, ‘Reform Jews’, ‘Reconstructist Jews’ and so on, but they deny that any but the first can correctly be said to be Jews.
In slightly more detail: A social context, as I am using the term, is a social structure—a family, friendship, school, hospital or court, an educational, medical, or legal organization or system, a town, state, or country, for example—relative to which people carry expectations (which typically produce norms) which shape behavior. My intention is that ‘social context’ picks out roughly the structures it often picks out in the sociological literature; see, for example, Zussmann et al. 2009. A social situation is, well, it’s a social situation—a conversation, a party, the second day of a deposition, a family dinner, etc.—in which people who carry, fear, resist, and so on various expectations about behavior interact. Situations, being concrete, tend to be contexts writ small; contexts, being temporally and spatially dispersed ways of interacting socially, are situations writ big.
Khanna and Johnson 2010 have a useful discussion of this and the notion of passing as having a racial identity in particular contexts.
The conditions under which one can acquire a social identity not assigned from the get go are likely quite different across different social identities. That said, I think the conditions under which such identity acquisition is possible have enough in common to justify posing a single question.
Exhibiting this desire and having it recognized by some Quakers appears basically all it takes to become a Quaker: see https://quaker.org/becoming-a-quaker/ (accessed September 2022).
Of course there is typically weighting involved, as some expectations will be much more important than others.
Here I borrow with some modification Cristina Bicchieri’s account of what distinguishes a social norm from such things as customs and descriptive norms. See Chaps. 1 and 2 of (Bicchieri 2017).
It is of course true that the Reconstructionist, if they are of Jewish ancestry, will be criticized for straying from Orthodoxy.
With the complication that (binary) gender identities, unlike religious identities, are dependent on one another. The norms to which men are expected to adhere and those to which women are expected to adhere are quite literally interdependent, norms for each identity involving norms for how to behave towards those with the other identity.
While the story I am trying to tell applies, I think, to gender identities as a whole, including non-binary identities, I’m going to ignore the extra complications such identities introduce, save to say that the last sentence of this paragraph should be qualified to make clear that since all gendered identities are in the relevant sense co-dependent (non-binary identities being in the first instance reactions to and rejections of the binary alternatives), it’s not just ‘pretty much everybody’ who is an ‘authority’ about what gender one is—everyone is.
There are affinities between the account of gender kinds in this section and the account in Dembroff (2018). Dembroff works with a notion of social context somewhat like the notion I work with here: Dembroff’s contexts are ‘communities of persons with shared clusters of beliefs, concepts, and attitudes that give rise to concrete social practices and structures’ (Dembroff 2018, p. 18); mine are standing and transient social relations in which there are expectations about behavior which potentially generate norms. I distinguish ‘concrete social situations’—roughly, situations in which members of a social context who are in close spatio-temporal proximity (or otherwise are in a position to directly impact one another’s behavior)—from social contexts. As I understand Dembroff, much of their discussion of social contexts is best understood as a discussion of concrete social situations.
Dembroff and I agree that ‘because gender kinds are socially constructed, we should expect operative gender kinds to vary across contexts’ (Dembroff 2018, p. 39). We disagree about the conditions under which self-ascriptions of gender identity are correct. Dembroff (tentatively, I take it) endorses Joshua Glasgow’s ‘modest pluralism’ about gender kinds, on which a trans person is a invariantly a member of many gender kinds across social contexts, though the relevance of such membership shifts across concrete social situations. I do not endorse this, as I think whether one is a member of a social(ly constructed) kind turns on whether the norms which determine that kind apply to one; I think this varies with the concrete situation one finds one’s self in. Put crudely, my view is that whether one has a social identity is in large part a matter of whether one is allowed to have it; as I understand Dembroff, their view is that one can correctly claim an identity so long as there is a context in which one would be allowed to claim it.
Dembroff writes that the ‘interesting and important project…is not asking whether a gender classification is true. Rather, it is determining what gender kinds operate in a social context, and stating their relationship to power and privilege’ (Dembroff 2018, p. 42). I agree that their project is important. But I do think that just as important is the question of when one can correctly ascribe a gender identity to oneself. If I am correct that the gender norms operative in a concrete situation control what one can say about one’s gender identity in that situation, it’s not clear that at the end of the day there is all that much difference in the projects we think are interesting and important.
I wish to remain non-committal about both the syntactic and the semantic mechanism that achieve this sort of thing, both in the case of ascriptions of fame and the case of gender ascription.
Aka ‘internal realism’; by the late 1980’s (Putnam 1987) he expressed mild regret that he did not initially call the view ‘pragmatic realism.’
I’m open to using a thicker notion of concept for present purposes.
(Putnam 1987, p. 19). The passage in question has to do with the ‘logical primitives’, in particular the idioms of quantification. But the point carries over pretty generally—on the same page Putnam argues that it carries over to notions like the notion point on a plane. (Putnam, by the way, uses the phrase ‘our notions’ in the passage cited above, not ‘our concepts’).
Putnam’s distinction between a concept and its uses and his insistence that it’s not possible to say what objects there are before we adopt a conceptual scheme—which seems to be a matter of settling on ways of using concepts—suggests, and maybe entails, that it is only concepts as they are used in a particular way that can be said to have extensions.
On the way I’m understanding Putnam, then, a conceptual scheme is a collection of concepts ‘being used’ in particular ways.
I am ignoring a lot of delicate issues of Putnamian hermeneutics here—in particular issues having to do with the relations, on Putnam’s view, between truth and justification, as well as the issue raised in Sosa (1993), as to whether Putnam’s view is consistent with the idea that there are sorts of things unrecognized by our conceptual scheme. I think for present purposes we can ignore those issues; I take up Sosa’s challenge in work in progress.
I’m adding something to our example here, and there are of course different views open to us about what would make a trans woman a woman. I think the argument here isn’t terribly sensitive to such details. (For a few relevant comments, see note 23.)
Note that I did not say that she knows that the conditions which make her claim true obtain; I said that she knows (of the conditions which make her claim true) that they obtain. We don’t require, in order to say that someone knows that p, that they know of conditions which make it true that p that they both know that such conditions obtain and that they are conditions that make it true that p. It generally suffices that they know of truth making conditions for p that the conditions obtain.
This needs a fair bit of i-dotting and t-crossing, of course; in particular, most of us don’t think that knowing that Twain is self-identical suffices to know that Twain is Clemens. This sort work on i’s and t’s is orthogonal to the issues we are discussing.
Objection: Mika says that the concept woman is a concept which can include trans women. You say Mika speaks truly, and that she is absolutely correct. But Alexa says that the concept is one which cannot include trans women. Don’t they speak truly, at least given their way of using ‘women’? Surely you must admit that they do. But then why think Mika is ‘absolutely correct’? Response: The fact that someone applies a concept in a particular way does not guarantee that the application is correct. In this case, Alexa is making a factual mistake. It is obvious that they are making a mistake: Mika correctly self-ascribes the concept woman. When she does so, she is using the concept in a way it can be used. Observation: It might be illuminating to compare Alexa’s view of the concept woman with the view of those who, some 40 or so years ago, objected that it is literally impossible for two men to be married. That was simply a mistake. Obergefell v. Hodges (the U.S. Supreme Court decision striking down laws barring same sex marriages) did not involve the introduction of a new concept, better expressed with a new word like ‘smarriage’ than with the word the Court used, ‘marriage’. The decision was a decision about marriage.
I’m grateful to a referee for pressing me to address this worry.
I want to say something about relations between the uses I make of Putnam’s view in this section and Jacob Hale’s fascinating discussion of the concept woman in ‘Are Lesbians Women?’ (Hale 1996).Â
Drawing on Bornstein (1994), Hale argues that the dominant concept of woman in the United States towards the end of the last century is a complicated cluster concept not unlike Harold Garfinkel’s ‘natural attitude’ towards gender. (Garfinkel 1967). According to Hale, it involved a large collection of characteristics and ways of being in the world, including biological characteristics (e.g., lacking a penis, having breasts), identifying as a woman, broadly cultural characteristics (e.g., engaging in ‘womanly pursuits’), and characteristics having to do with maintaining a culturally recognized ‘female presentation’. Hale argues that none of these characteristics is necessary or sufficient for being a woman—varying ways of satisfiying one or another cluster can suffice—but that the dominant late ‘90s U.S. conception weighs the sexual ones much more heavily than the others.
I think we ought to take something close to Hale’s view on board; indeed, I think we should say that something like Hale’s elaboration of ‘the natural attitude’ towards gender in a certain sense constitutes our common conception of gender kinds. Western conceptions of woman and men (that is, particular ways of conceptualizing women and men and applying gender terms like ‘woman’, ‘’man’, ‘Frau’, ‘Mann’,‘mujer’ and ‘hombre’) are not united in embracing and employing all of the characteristics Hale lists (as well as a corresponding collection of ‘natural markers’ of being a man or a woman). Rather, these conceptions are in good part united by the recognition that something like Garfinkel’s natural attitude towards gender (was and) is the standard way of conceptualizing women and men: to have the concepts man and woman is in good part to recognize that, according to the dominant way of conceiving these kinds, one is supposed to conform to the dominant attitude towards gender. (I develop a picture of conceptualization and meaning along these lines in Richard (2019).) To think of gender conceptions in this way is at once to see them as ur-produced by a culturally inculcated way of looking at gender while also acknowledging that our conceptions of gender can and often do resist buying into the natural attitude towards gender. Appropriating Putnam’s pragmatic realism seems to me to be as good a way of doing justice to this as any.
Or, if you’re on the other side of the contestation, to resist such change.
The referee mentioned in note 23 observes that it is not just the truth of Mika’s self-ascription of gender that should concern us. Alexa speaks as a member of groups whose voices are dominant; Mika speaks as a member of groups which are minoritarian. The referee asks what exactly makes a scheme dominant or minoritarian, and how contestation between schemes is and ought be resolved.
As the referee also observes these are huge questions, and I can’t address them here. I will say that the first question—what makes a scheme dominant—is to my way of thinking as much sociological as philosophical. That said, it behooves philosophers who see conceptual engineering as a vehicle for social engineering to look more closely at the nature and dynamics of social contestation. Practically any worthwhile attempt to modify the meaning of a philosophically or socially important term will meet with resistance; an understanding of the details of past successful (and unsuccessful) contestation of social norms and ideology is necessary for theorizing about how to change dominant ways of thinking. (Chap. 4 of Richard (2019) has some discussion of how one might describe about the dynamics of public language meaning change, but even the author of that work would admit that an awful lot more needs to be said.) As for the normative question—how ought contestation be resolved—well, as they say, this is a topic for further research.
References
Bicchieri C (2017) Norms in the wild: how to diagnose, measure, and change social norms. Oxford University Press
Borenstein K (1994) Gender Outlaws: on women, men, and the rest of us. Routledge
Cappelen H (2018) Fixing language. Oxford University Press
Dembroff R (2018) Real talk on the metaphysics of gender. Philos Topics 46(2):21–50
Garfinkel H (1967) Studies in ethnomethodology. Polity Press
Hale J (1996) Are lesbians women? Hypatia 11(2):94–121
Jackman H (1999) We live forwards but understand backwards: linguistic practices and future behavior. Pacific Philos Q 89:157–177
Khanna N, Johnson C (2010) Passing as Black: racial identity work among Biracial Americans. Social Psychol Q 73:380–397
Putnam H (1987) The many faces of realism. Open Court
Richard M (2019) Meanings as species. Oxford University Press
Sosa E (1993) Putnam’s pragmatic realism. J Philos 90:605–26
Zusman M, Knox D, Gardner T (2009) The social context view of sociology. Carolina Academic Press
Acknowledgements
Thanks to Nancy Bauer, Tim Sundell, Giulia Terzian, Quinn White, a referee for Topoi, and audiences at the Metalinguistic Disagreement and Semantic Externalism conference at Nova University, Lisbon, Portugal and an APA symposium on conceptual engineering for helpful comments.
Funding
No funds, grants, or other support was received for this work.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Richard, M. Conceptual Engineering: Be Careful What You Wish for. Topoi 42, 1063–1073 (2023). https://doi.org/10.1007/s11245-023-09928-z
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11245-023-09928-z