![Sample our Humanities journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5939/TOC-Subject-Sample-2021-Humanities.jpg"})
ABSTRACT
Mary Kate McGowan's Just Words offers an interesting account of exercitives. On McGowan's view, one of the things we do with words is change what's permitted, and we do this ubiquitously, without any special authority or specific intention. McGowan's account of exercitives is meant to identify a mechanism by which ordinary speech is harmful, and which justifies the regulation of such speech. It is here that I part ways. I make three main arguments. First, McGowan's focus on harm is misguided; that ordinary speech is harmful is harder to support, and in turn does not less to support conclusions about whether it is wrong and warrants regulation, than Just Words suggests. Second, if the speech is harmful, McGowan's account of exercitives is ill-suited to explain why; the relevant instances of ordinary speech seem to express speakers' views about moral norms, not change social norms. Third, McGowan's argument for why ordinary speech should be regulated because it is a form of discrimination presupposes a great deal that is controversial about what makes discrimination wrong and legally actionable, and even if it is right it may license the regulation of more speech than many, including McGowan, would deem acceptable.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 Austin (Citation1962, 150). Austin’s initial list of examples was ‘appointing, voting, ordering, urging, advising, warning’ (150), with a longer list appearing later in the book (154–155).
2 Austin (Citation1962, 154).
3 This example is McGowan’s (Citation2019, 58).
4 See, inter alia, Green (Citation1998) and Saul (Citation2006).
5 Of the seven chapters in the book, the first three offer the account of conversational exercitives, and the fourth generalizes this into an account of covert exercitives.
6 McGowan (Citation2019, 2).
7 Id.
8 McGowan (Citation2019,, 3).
9 Id. The view is not that speech regulation is all things considered justified. But it’s not completely clear what this weaker claim amounts to. As Berman recently noted, ‘‘warranted’ can be a waffle when not clearly defined in terms of more familiar normative notions as ‘good,’ ‘reason,’ ‘duty,’ and ‘ought’’ (Citation2023, 10, n. 16).
10 McGowan (Citation2019: 1, ch. 5).
11 Others have typically defended this view by arguing for the legal regulation of hate speech. McGowan contrasts the argumentative strategy in Just Words with other approaches in ch. 7 (see especially 163–164).
12 McGowan (Citation2019, 1).
13 Mill (Citation1859: ch. 3) took this to be so with regard to claims that corn dealers starve the poor.
14 The examples I’ve considered most involve generics, gender pronouns, and ordinary racial attributions: Wodak, Leslie, and Rhodes (Citation2015), Wodak and Leslie (Citation2017), Dembroff and Wodak (Citation2018; Citation2021), and Wodak (Citation2022).
15 I appreciate that this second point may sound like I’m splitting hairs, at least until I delve into the details of McGowan’s view. But to at least add some presumptive force to the point, compare Maitra, who distinguishes speech that constitutes subordination from speech that causes ‘its hearers to form beliefs that, in turn, cause those hearers to treat others in ways that subordinate them’ (Citation2012, 97). If my diagnosis is right, ordinary sexism turns out to be an instance of the latter, not speech that constitutes subordination or constitutions harm.
16 McGowan (Citation2019, 26).
17 Lewis (Citation1983). As McGowan notes (Citation2019, 31, 97fn. 11), Lewis framed this in terms of rules rather than norms.
18 McGowan (Citation2019, 31).
19 McGowan (Citation2019, 34).
20 McGowan (Citation2019, 45).
21 McGowan (Citation2019, 90).
22 For example, McGowan holds (as I noted above) that it matters whether we abide by g-norms (see, e.g. 109–110). But it’s unclear what it takes to abide by norms, and whether this will be consistent with McGowan’s views that such exercitives do not require speakers to have specific intentions. See Mikkola (Citation2021, 466) on this issue.
23 McGowan (Citation2019, 109).
24 McGowan (Citation2019, 110).
25 McGowan (Citation2019, 111).
26 McGowan (Citation2019, 112). McGowan treats this claim as a ‘hypothesis.’
27 McGowan (Citation2019, 112).
28 Though it’d require McGowan to give up on the official account of how an utterance can constitute harm, which requires three things: ‘The utterance enacts a norm; that norm is followed and harm results from following that norm’ (Citation2019, 24). This makes the harm of the utterance contingent on its results.
29 McGowan (Citation2019, 102).
30 McGowan repeatedly appeals to ‘the harm of discrimination’ (e.g. Citation2019, 24), and then later explains why it is harmful as follows: it is a ‘relative harm with an implicit comparison class. To say that non-whites are discriminated against in the United States today, for example, is to say that they are disadvantaged relative to whites in the United States today’ (167 n. 24, emphasis in original).
31 This is the counterfactual account. Similar issues arise for most comparative accounts of harm. And as Rasmussen (Citation2019) argues, prominent non-comparative accounts of harm provide a poor basis for explaining the wrong of discrimination.
32 As Maitra and McGowan (Citation2021, 325) note, ‘There are different accounts of harm in the philosophical literature.’ There, and here, McGowan aims to remain fairly neutral between them. McGowan seems to understand harm in terms of results (Citation2019, 24), though perhaps not ‘bad effects’ (176: 44).
33 Anderson (Citation1999, 289).
34 Rasmussen (Citation2019, 874). Rasmussen goes on to argue that harm can be both a defining and wrong-making feature, but only if we adopt a ‘non-orthodox’ account of harm.
35 Hellman (Citation2008, 7–8).
36 Hellman (Citation2008, 8). Similarly, for Eidelson, ‘discrimination is morally troubling not simply because it gives rise to harm or unfairness, but because it manifests a form of basic disregard for the standing as persons of those who are discriminated against’ (Citation2015, 126).
37 Defenders of harm-based views accept this contingency. See Kasper Lippert-Rasmussen (Citation2006, 174).
38 Though they may be rivals, in the following sense. If the relevant speech is only harmful in virtue of being degrading (or oppressive, or …), and the degradation (or …) is already sufficient for the speech to be wrong, I don’t think harm can explain why the speech is wrong (for the reasons I offer in Wodak Citation2020). But harm could just be an aggravating factor – it could help make the speech more seriously wrong.
39 Langton (Citation1993).
40 This comes up often. For example: McGowan ‘aims to work within an actual free speech system (that of the U.S.) rather than argue for a new one’ (Citation2019, 181).
41 See Gardner for an illuminating discussion of (in relation to discrimination) of how conflating these issues can yield ‘egregious mistake[s]’ (Citation1996, 365–367).
42 See Feinberg (Citation1984).
43 For Rawls (Citation1972), we can legally regulate conduct to ensure justice is done. Gardner notes that this Rawlsian view and Millian doctrines about harm-prevention are ‘cross-cutting’ (Citation1996, 365).
44 Much of U.S. free speech jurisprudence has been in the grips of the idea that a free marketplace of ideas is an optimal institution for promoting true belief, which Goldman and Cox (Citation1996) note betrays a profound misunderstanding of the economic theory of what marketplaces maximize.
45 See Schauer’s illuminating discussion of how ‘much public rhetoric, academic commentary, and even legal doctrine seems to often deny’ obvious truths about how speech can be harmful (Citation2011, 81).
46 For one thing, Steve’s utterance would also make it conversationally appropriate for John to reject the use of degrading terms for women. For another, McGowan ties oppression to broader social norms.
47 McGowan (Citation2019, 111). We’re also told that the case involves changes to ‘harmful social norms’ (125).
48 McGowan (Citation2019, 25); Bicchieri (Citation2006).
49 For some examples, see McGowan (Citation2019, 69 n. 15, 88, 101).
50 McGowan (Citation2019, 112).
51 Bicchieri puts this in terms of conditional ‘commitments’ or ‘preferences’ (Citation2006, 20). Brennan et al (Citation2013) similarly demarcate social norms in terms of whether reasons are conditional (or ‘independent’).
52 McGowan could reject this view, but aside from being contentious and undefended, this’d risk giving us an overly broad account of social norms. For example, one way in which the conditional nature of social norms is important is that it helps distinguish them from collective habits. Bicchieri makes this point with the example of wearing warm clothes in winter (Citation2006, 21–22): we might expect others to do this, descriptively and normatively, but that doesn’t make it a social norm because our expectations aren’t conditional on others in the right way. I should note that at times McGowan’s writing suggests some background assumption that any norms that govern social practices are social norms (e.g. 111). But that is inconsistent with Bicchieri’s view, and seems as dubious as the view that any norms that govern conversations are conversational norms. (Can’t moral norms govern both?)
53 For what it’s worth, I think this same issue arises with McGowan’s first case of a non-conversational covert exercitive (Citation2019, 90–91). The example involves the speaker declaring that they are happy for Paul, to an audience who knows Paul bullied the speaker back in high school. McGowan writes that this is a move in a conversation and ‘a move in the norm-governed activity of social interaction.’ But not all of the norms that govern social interaction are social norms! It is plausible that the speaker’s ‘magnanimous’ way of talking about Paul ‘encourages taking the high road’, but does it do so via changing social s-norms, or via changing others’ personal moral norms?
54 Interestingly, in discussing CitationLangton and West’s position on pornography (Citation1999), McGowan switches to whether pornography says women are ‘socially inferior’ (Citation2019, 132), rather than just ‘inferior’, and of what’s ‘socially legitimate’ rather than just ‘legitimate’. This makes the point made above easy to miss. Subordinating women ranks them as morally inferior, not as socially inferior in some non-moralized sense.
55 I’m assuming here that social norms are constellations of normative judgments, but moral norms aren’t. So if Steve’s utterance makes others adopt sexist moral beliefs, it doesn’t thereby make it morally permissible to treat women in sexist ways. That’s why on my diagnosis it doesn’t enact s-norms – my talk about changes in personal norms is needn’t mean there are any changes in what morality permits.
56 McGowan doesn’t say this, but does agree that not all norm change is norm enactment (Citation2019, 19).
57 Steinhoff (Citation2022) makes a similar criticism of the example. Worse yet, if these norms are so hard to identify, it’s unclear why we should think Steve is abiding by them, which McGowan’s account requires.
58 McGowan (Citation2019: ch 6). The relevant ‘permissibility facts’ enacted are again meant to be social (e.g. on 141–142). A further note about McGowan’s claims here, which I owe to Eleonor Neufeld. McGowan is concerned with cases of publicly rather than privately consuming pornography – e.g. hanging a lewd poster at work where it’ll be seen by others (136), rather than surreptitiously viewing the same image at home. This is a subtle change of focus from much work on how pornography silences and subordinates.
59 McGowan does not claim that such speech should all things considered be regulated, noting the ‘substantial practical challenges’ that would need to be overcome for such a position to be defended (181).
60 McGowan (Citation2019, 164).
61 McGowan (Citation2019, 169, 173).
62 Wilson (Citation1974, 5). Schauer provides a helpful overview of the history of the precept, and authors who’ve made this same point (Citation2018, 437–438).
63 McGowan (Citation2019, 1).
64 McGowan (Citation2019, 165).
65 McGowan (Citation2019, 167).
66 Hellman (Citation2016, 895).
67 Hellman (Citation2016, 896).
68 See, among others, Hellman (Citation2008); Lippert-Rasmussen (Citation2013); Eidelson (Citation2015); Khaitan (Citation2015); Solanke (Citation2017); Moreau (Citation2010, Citation2020); and Hellman and Moreau (Citation2013).
69 Watson (Citation2021, 533). Watson says that this is true ‘under some of the relevant civil rights laws’, but not all.
70 Hellman (Citation2016, 899).
71 McGowan (Citation2019, 167, n. 24).
72 Moreau (Citation2020, 210).
73 Watson (Citation2021, 532–535) offers a great discussion of this issue for McGowan’s argument.
74 The label comes from Siegal (Citation2004), but the distinction was first raised in some form by Fiss (Citation1976).
75 The label and definition are from Siegal (Citation2004: 1472–1473).
76 Bostock v. Clayton County, 590 U.S. ___ (2020). This point is made by Berman and Krishnamurthi (Citation2021) and Eidelson (Citation2022).This article was original submitted in August 2022, prior to Students for Fair Admissions v. Harvard, 600 U.S. 181 (2023).