Conceptual engineering and the dynamics of linguistic intervention

ABSTRACT

The Implementation Problem for conceptual engineering is, roughly, the problem conceptual engineers face when attempting to bring about the conceptual change they support. An important aspect of this problem concerns the extent to which attempting to implement concepts can lead to unintended negative consequences. Not only can conceptual engineers fail to implement their proposals, but their interventions can produce outcomes directly counter to their goals. It is therefore important to think carefully about the prospect of attempted implementation leading to unintended negative consequences: what sort of negative consequences can conceptual engineers expect? Are some forms of conceptual engineering more likely than others to lead to such consequences? And is conceptual engineering still viable even given the risk of such consequences? This paper addresses such questions. I begin by outlining different forms of conceptual engineering (Section 2), before examining how they can produce unintended negative consequences (Section 3). I then discuss some implications of the fact that attempted implementation can produce unintended negative consequences, suggesting that, among other things, some forms of conceptual engineering are less viable than others (Section 4). I conclude, though, by considering some of the ways in which conceptual engineering is nonetheless a worthwhile pursuit (Section 5).

KEYWORDS:

• Concepts
• conceptual engineering
• the implementation problem
• linguistic intervention
• unintended consequences

Acknowledgements

Thanks to Louise Antony, Alex Guerrero, Jennifer Nado, Eleonore Neufeld, Gary O'Brien, and several anonymous referees for their feedback on this paper.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 For helpful overviews of many important issues related to conceptual engineering, see Cappelen (2018), Cappelen and Plunkett (2020), Isaac (2021), Koch (2021), Isaac, Koch, and Nefdt (2022), and Koch, Löhr, and Pinder (2023).

2 Neufeld (forthcoming) refers to this problem as the Feasibility Question rather than the Implementation Problem. (See also Machery (2021)). I prefer ‘implementation problem’ given its previous usage in the literature, but nothing hangs on this terminological point.

3 See Löhr (2022) for earlier work focusing on ethical issues related to linguistic interventions and conceptual engineering. See also Sterken (2020).

4 Of course, similar claims have been defended by other philosophers (Chalmers 2020; Koslow 2022). However, as I show in Section 3, my analysis diverges from theirs in certain notable respects.

5 How exactly these unintended consequences arise and how they impact such forms of conceptual engineering are matters I leave for future investigation.

6 See Silver and Shaw (2022) for some discussion of how failing to take sides on certain controversial issues can lead one to be viewed negatively by others.

7 For some examples of this large and growing literature, see Plunkett and Sundell (2013), Simion and Kelp (2019), Chalmers (2020), Thomasson (2020), Löhr (2021), Nado (2021), Riggs (2021), Jorem (2022), Jorem and Löhr (2022), and Isaac (2023).

8 See Chalmers (2020) and Koslow (2022) for similar remarks.

9 On dogwhistles, see Saul (2018).

10 Cf. Chalmers (2020, 9–12).

11 See Fricker (2007) for relevant discussion.

12 See Guerrero (2014) for a classic outline and defense of lottocracy.

13 For criticism of his arguments, see Pepp, Michaelson, and Sterken (2019) and Brown (2019). For a response, see Habgood-Coote (2022).

14 One might also include concept preservation as a form of conceptual engineering, together with its own distinctive linguistic interventions (i.e., striving to use certain expressions even as their wider use declines, exhorting or otherwise incentivizing others to use them, and so on). See Lindauer (2020) for relevant discussion.

15 Note that I am not claiming that people must understand conceptual proposals in order to accept them. After all, people may simply defer to the conceptual proposals of others. But sincere agreement with the proposals (and subsequent acceptance) plausibly requires an understanding of them. For related discussion of semantic deference, see Pollock (2019).

16 Presumably one would want to make further precisifications, not least in relation to other concepts such as LIBERALISM and DEMOCRACY which are themselves often heavily contested. I set aside such complications here. On neoliberalism more generally, see Vallier (2022).

17 On politically motivated reasoning, see Lodge and Taber (2013).

18 On lexical effects, see Cappelen (2018, 122–134).

19 Two comments are in order. First, by foregrounding this possibility, I do not mean to suggest that conceptual engineers themselves are incapable of such self-interested behavior. (See Shields (2021) for relevant discussion.) Second, I also do not mean to suggest that one cannot correctly use such concepts to further one’s own ends. However, in what follows I primarily have in mind cases where people deliberately misapply the relevant concept.

20 See Gibbons (2023) for relevant discussion.

21 For instance, see Tirell (2012) on linguistic violence.

22 This is not to suggest that something like this could never be traumatic even by the standards of more typical, less expansive concepts of TRAUMA. Much would depend on the precise details of any given case.

23 Some of the latter group may even want to contract the relevant concepts, such that, for example, actions or events commonly thought of as violent no longer fall under the scope of the concept, experiences commonly thought of as traumatic are no longer viewed as traumatic, and the like. But such efforts would face similar problems to efforts to expand those concepts. Conceptual contraction, much like conceptual expansion, can lead to unintended and undesired consequences.

24 Cf. Löhr (2022, 843).

25 As an example of this sort of response, consider Suzanna Danuta Walters’ comments in the aftermath of the scandal involving Rebecca Tuvel’s article ‘In Defense of Transracialism’ (Tuvel 2017). Walters wrote that ‘the idea that any article in a specialized feminist journal causes harm, and even violence … is a grave misuse of the term “harm”’ (Walters 2017). Commenting on the same scandal, José Luis Bermúdez claimed that ‘the concept of harm has been twisted beyond all recognition’ (Bermúdez 2017).

26 See Case (2019) for relevant discussion.

27 It should be noted that Koslow also recognizes the limitations of such strategies, writing that ‘few engineering proposals with a revolutionary flavor will involve broadening or narrowing in this sense’ (Koslow 2022, 19).

28 On representational complacency, see Cappelen (2018, 5–7). See also Belleri (2023) for related discussion.

29 Other philosophers have also commented on the possible usefulness of case studies. For example, see Gibbons (2022, 19–20) and Löhr (2022, 845). See also Landes (forthcoming) for related discussion.