Introduction

Political exclusion does not always take the form of explicit slurs or openly discriminatory propositions. Political speakers may instead use expressions that support different interpretations for different audiences: an apparently innocuous interpretation may be available to a wider audience, while a politically informed subgroup recovers an additional coded meaning. Such utterances are often described as “dog-whistles.” Dog-whistle communication is particularly suitable for Clark’s “Exploring Implicatures” task because the central question is not only what an utterance explicitly communicates, but also what further assumptions hearers are encouraged to derive, and why these assumptions may be more accessible to some audiences than to others. This report examines four utterances drawn from Japanese right-wing political discourse and asks whether and how they give rise to anti-LGBTQ+ interpretations. Relevance Theory is used to explain how differences in contextual accessibility allow different audiences to derive different interpretations of overt dog-whistles, while Speech Act Theory is used to account for the perlocutionary effects of covert dog-whistling.

Saul (2018) distinguishes four types of dog-whistle according to two dimensions: dog-whistles may be overt or covert, and intentional or unintentional. This report focuses primarily on intentional dog-whistles. In an overt intentional dog-whistle, a speaker deliberately uses an utterance with a coded interpretation intended for a subgroup of the general audience, while the wider audience remains unaware that the second interpretation exists. The target subgroup, by contrast, consciously recognises the coded message as intended. Covert intentional dog-whistles operate differently: rather than necessarily communicating a particular coded proposition, they may bring pre-existing attitudes to salience and thereby influence the audience’s judgement outside conscious awareness. Saul ultimately analyses these as covert perlocutionary acts, whose intended effect depends on the audience not recognising that effect as intended.

Theoretical Framework

Overt Intentional Dog-whistles: Relevance Theory

Relevance Theory (RT) can be used to explain how overt intentional dog-whistles produce different interpretations for different audiences. RT treats an utterance as an ostensive stimulus: by speaking, the speaker attracts the audience’s attention and creates a presumption that the utterance is worth processing. Hearers consider interpretations in order of accessibility, following a path of least effort, and stop when an interpretation satisfies their expectations of relevance. Since different hearers possess different cognitive environments, the contextual assumptions available to them may differ, allowing the same utterance to yield different optimally relevant interpretations. A well-known example discussed by Saul is George W. Bush’s utterance in his 2003 State of the Union speech:

Yet there’s power, wonder-working power, in the goodness and idealism and faith of the American people.

For a non-fundamentalist listener, wonder-working power may be processed simply as conventional political or religious-sounding rhetoric. This interpretation may already provide sufficient relevance, so no further group-specific interpretation is required. For a fundamentalist Christian listener, however, wonder-working power is associated with the power of Christ. This group-specific knowledge makes an additional interpretation highly accessible: Bush is invoking a specifically Christian expression and thereby signalling affinity with fundamentalist Christians. Saul notes that the phrase may communicate either a more explicitly Christian interpretation or simply the fact that Bush “speaks their idiolect.”

Covert Intentional Dog-whistles : Speech Act Theory

In Saul’s account, covert intentional dog-whistles are analysed using Austinian speech-act theory, specifically as a special type of perlocutionary act. Perlocutionary effects are consequences that an utterance has on an audience’s thoughts, feelings, or actions. Unlike overt intentional dog-whistles, covert intentional dog-whistles do not necessarily involve the recovery of a particular communicated proposition. Instead, they may work by making pre-existing attitudes or associations more salient and thereby influencing the audience’s judgement.

Saul calls these covert perlocutionary acts. Their defining property is that the speaker intends to produce a particular effect on the audience, but the effect will not succeed if the audience recognises that effect as intended. For example, a speaker using a covert racist dog-whistle may intend to activate racial resentment or stereotypes without wanting the listener to recognise: “This speaker is deliberately trying to activate my prejudice.” Once the manipulative intention becomes consciously recognised, the intended perlocutionary effect may be neutralised.

Examples and Analysis

This section analyses four examples of potentially dog-whistling expressions in Japanese right-wing political discourse, all drawn from Yamaguchi et al. (2023).

1.「伝統的家族を守る」(protect the traditional family)

Background: Constitutional revision has long been an important political objective for conservative and right-wing politicians in Japan. One proposed revision concerns Article 24 of the Constitution, which establishes principles of individual dignity and equality between the sexes in matters relating to marriage and the family. Some conservative politicians have advocated revising this provision in order to give greater constitutional recognition to the traditional family and to promote the idea of 「伝統的家族を守る」 (“protecting the traditional family”). (Yamaguchi et al., 2023, pp. 187-189)

Analysis: This case can be analysed as a typical case of an overt intentional dog-whistle. The utterance itself makes no explicit reference to homosexuality, same-sex marriage, or LGBTQ+ people. For a general audience, accessible contextual assumptions may include the idea that families are socially valuable and that politicians commonly seek to support family life. The expression may therefore be interpreted simply as expressing a commitment to preserving or supporting family life. For a right-wing target audience, however, 「伝統的家族」 (“traditional family”) can activate a more specific set of contextual assumptions in which the normative family is based on heterosexual marriage and conventional gender roles, and is contrasted with same-sex and other LGBTQ+-inclusive family forms. In this cognitive environment, 「伝統的家族を守る」 can yield the additional implicated conclusion that same-sex marriage and other LGBTQ+-inclusive understandings of family should be resisted. The verb 「守る」 (“protect”) further encourages the hearer to search for a potential threat from which the traditional family requires protection; in the relevant political context, LGBTQ+-inclusive reforms provide a highly accessible candidate for that threat.

2.「すべての国民が安心して生活することができるよう、留意する」(due consideration shall be given so that all citizens can live with peace of mind)

Background: In the context of the Tokyo Olympics and the 2023 G7 Hiroshima Summit, Japan faced increasing international pressure, particularly from Western governments, to strengthen legal protections for LGBTQ+ people. Against this background, the Liberal Democratic Party (LDP) led government enacted the so-called LGBT Understanding Promotion Act (LGBT理解増進法). During the legislative process, however, disagreements emerged between proponents of stronger protections and conservative politicians within and outside the LDP. The final text therefore incorporated several amendments intended to accommodate conservative concerns. One of these was the provision that “due consideration shall be given so that all citizens can live with peace of mind” (「すべての国民が安心して生活することができるよう、留意する」). (Yamaguchi et al., 2023, pp. 208-213)

Analysis: This case can also be analysed as a typical example of an overt intentional dog-whistle. For a general audience, accessible contextual assumptions may include the ideas that legislation should take the interests of all citizens into account and that the government should avoid unintended negative consequences. The utterance may therefore be interpreted simply as meaning that “the law should take everyone’s welfare into account.” Since this interpretation already satisfies the hearer’s expectations of relevance, there is no need to search for an additional interpretation. For an anti-LGBTQ+ target audience, however, further contextual assumptions may be highly accessible, such as the beliefs that LGBTQ+ rights conflict with conservative social values or that LGBTQ+ inclusive education threatens children. Given these contextual assumptions, the same utterance can yield an additional implicated conclusion: LGBTQ+ inclusion may threaten the security or peace of mind of other citizens and should therefore be limited or constrained. The wording thus allows the general audience to recover an innocuous interpretation while enabling the target audience to recognise a more exclusionary coded message. Importantly, LGBTQ+ people are themselves semantically included in the literal meaning of 「すべての国民」 (“all citizens”). Pragmatically, however, the wording may encourage the target audience to contrast LGBTQ+ people with other citizens whose 「安心」 (“peace of mind”) is presented as potentially threatened. This contrast therefore arises from contextual inference rather than from the literal meaning of the expression.

3.「Children First」

Background: In April 2021, the Ministry of Education, Culture, Sports, Science and Technology (MEXT) began the phased introduction of “life safety education” (生命の安全教育) in primary and secondary schools nationwide. The programme included material relating to sexuality and basic knowledge of LGBTQ+ issues. These developments prompted criticism from conservative politicians within the LDP as well as from right-wing groups outside the party. Against this background, LDP politicians established a study group under the slogan “Children First.” (Yamaguchi et al., 2023, pp. 214-215)

Analysis: Again, this case can be analysed as an overt intentional dog-whistle. For a general audience, accessible contextual assumptions may include the idea that the government should protect children’s welfare and provide appropriate education. The slogan Children First can therefore be understood simply as expressing the principle that children’s interests should be given priority. Importantly, however, the word first introduces a priority relation and therefore invites the hearer to consider which other interests are being assigned lower priority. For the target audience, the preceding controversy over LGBTQ+ inclusive education makes LGBTQ+ rights and inclusion a highly accessible candidate for such a competing interest. The audience can therefore derive the additional implicated conclusion that LGBTQ+inclusive education conflicts with children’s interests and should be restricted in order to protect children. The slogan thus leaves the competing interest unspecified for the general audience, while allowing the target audience to identify LGBTQ+ inclusion as an interest from which children are presented as needing protection.

4. 「女性の権利保護」(protection of women’s rights)

Background: Following the passage of the aforementioned legislation promoting understanding of LGBTQ+ issues, debates over transgender rights increasingly focused on access to women’s spaces and the participation of transgender women in women’s sports. Conservative groups raised concerns about what were commonly framed as the “restroom issue” and the “female athlete issue,” while organizations such as the Association to Protect Women’s Spaces (女性スペースを守る会) were established. Within the LDP, similar concerns were articulated through calls for the 「女性の権利保護」 (“protection of women’s rights”). Notably, some of these arguments also received support from commentators identifying with feminist or left-wing positions. Articles critical of transgender-inclusive policies from the perspective of protecting women’s rights also appeared on the Women’s Action Network (WAN), a feminist academic platform associated with Ueno Chizuko. (Yamaguchi et al., 2023, pp. 219-227)

Analysis: This case is more complex than the previous examples because the same discourse may operate through several dog-whistling mechanisms. For a conservative target audience, 「女性の権利保護」 can function as an overt intentional dog-whistle. For a general audience, the expression may simply be interpreted as affirming that the government should protect women’s rights. For an audience familiar with trans-exclusionary discourse, however, additional contextual assumptions may be highly accessible: for example, that trans-inclusive policies conflict with the interests or safety of cisgender women. These assumptions allow the target audience to derive the additional implicated conclusion that protecting women’s rights requires resisting or limiting some forms of trans inclusion.

The same broader discourse may also produce covert perlocutionary effects on audiences who do not consciously recognise its anti-trans framing. Repeated emphasis on restrooms, women’s spaces and women’s sports may make concerns about safety, vulnerability and fairness more salient, leading some hearers who would not consciously endorse anti-trans discrimination to perceive trans-inclusive policies as threatening to women. In Saul’s terms, this would resemble a covert dog-whistle insofar as the relevant effect operates without the audience recognising the speaker’s intention to produce it. However, establishing that this is an intentional covert dog-whistle would require evidence that the original speakers deliberately sought such an effect on this audience.

A further possibility is especially relevant to Saul’s account of unintentional dog-whistles. Once members of other political groups adopt and repeat the “women’s rights versus trans rights” framing without intending its exclusionary effects, their utterances may function as what Saul calls “amplifier dog-whistles”: they reproduce and extend the effects of an original dog-whistle even though the immediate speakers do not share or recognise the original manipulative intention.

Discussion

A Recurring “Protection-Threat” Pattern

A common pattern emerges across the four examples. Each expression foregrounds a socially valued group, institution, or interest that supposedly requires protection: the traditional family, the security of all citizens, children, or women’s rights. None of these expressions explicitly identifies LGBTQ+ people as a threat. However, within the relevant political context, LGBTQ+ inclusive reforms provide a highly accessible candidate for the missing source of danger or competing interest. The dog-whistle therefore works not primarily by explicitly attacking LGBTQ+ people, but by constructing a protection–threat relation in which LGBTQ+ inclusion occupies the implicit position of the threat. What Relevance Theory Adds to the Analysis The first three cases, together with the overt dimension of the fourth, illustrate the usefulness of Relevance Theory for analysing dog-whistles.The coded interpretation does not have to be encoded in the expression itself. Instead, it arises from the interaction between the utterance and audience-specific contextual assumptions. A general hearer may reach an innocuous interpretation that already satisfies expectations of relevance, while a politically informed target audience has access to additional assumptions that make an exclusionary interpretation highly accessible. The asymmetry between audiences is therefore not simply a matter of semantic ambiguity, but of differences in their cognitive environments.

The Limits of an Implicature-Based Relevance Theory Analysis

The distinction between overt and covert dog-whistles also reveals a limit of an implicature-based analysis. Relevance Theory provides a strong explanation of overt cases because the target audience consciously recovers an additional speaker-intended interpretation. Covert dog-whistles, however, may operate by activating pre-existing fears, stereotypes or associations without the audience recovering a particular proposition. These cases are better captured by Saul’s notion of a covert perlocutionary act which derives from Austin’s speech act theory.

Methodological Limitations

Demonstrating that an utterance supports different interpretations for different audiences is not by itself sufficient to prove that the speaker deliberately designed it as a dog-whistle. Establishing intentional dog-whistling requires additional contextual evidence, such as the political history of the expression, repeated use within an ideological community, statements by political actors, or evidence that the expression was introduced specifically in response to conservative demands. This report therefore identifies expressions that are strongly compatible with Saul’s account of intentional dog-whistling, while recognising that speaker intention cannot always be established from the utterance alone. Similarly, the analysis reconstructs the contextual assumptions available to target audiences from the surrounding political discourse rather than directly testing audience interpretations. Reception studies, interviews, experimental data, or corpus evidence from target communities would therefore be needed to establish how consistently these coded meanings are actually recovered.

This report also only briefly addresses the other two types of dog-whistling identified by Saul: overt unintentional dog-whistles and covert unintentional dog-whistles. These are equally important because they show that the political effects of dog-whistling do not always depend on the intention of the immediate speaker. In particular, Saul discusses amplifier dog-whistles, in which speakers repeat expressions or frames originating in intentional dog-whistling and thereby extend their effects without intending or even recognising the original manipulative function. This is particularly important when dog-whistle expressions spread beyond the political groups in which they originally developed: a speaker may sincerely reject an exclusionary ideology while nevertheless reproducing associations created by it. Saul therefore notes that unintentional covert dog-whistles can turn speakers into “mouthpieces for an ideology that they reject.” A fuller analysis of these unintentional forms would require examining how dog-whistle expressions circulate, change meaning, and continue to produce perlocutionary effects after they leave the discourse of their original users. Because the present report focuses mainly on audience-specific implicatures and intentional political signalling, the development and circulation of unintentional dog-whistles remain topics for further research.

Conclusion

This report has examined four expressions from Japanese right-wing political discourse in order to show how dog-whistle communication can contribute to anti-LGBTQ+ exclusion without relying on explicitly discriminatory language. The first three examples—「伝統的家族を守る」, 「すべての国民が安心して生活することができるよう、留意する」, and Children First—can be analysed primarily as overt intentional dog-whistles. Their surface interpretations are compatible with widely accepted values such as supporting families, protecting citizens and prioritising children. For politically informed target audiences, however, additional contextual assumptions make more exclusionary interpretations accessible, including resistance to same-sex marriage and LGBTQ+-inclusive education.

Relevance Theory helps explain this audience asymmetry. The same utterance interacts with different cognitive environments, allowing a general audience to reach an innocuous interpretation while enabling a target audience to recover an additional coded implication. The case of 「女性の権利保護」 is more complex, since the discourse can function both through consciously recognised coded meanings and through the activation of concerns about safety, vulnerability and fairness. Such effects demonstrate the limits of analysing all dog-whistles purely in terms of implicature and support Saul’s use of perlocutionary acts to account for covert dog-whistling.

More broadly, the four cases reveal a recurring protection–threat frame in Japanese anti-LGBTQ+ discourse. Rather than explicitly presenting LGBTQ+ people as undesirable, these expressions foreground the family, citizens, children or women as groups requiring protection, leaving LGBTQ+ inclusion to occupy the implicit position of the threat. Dog-whistle communication can therefore make exclusionary political positions communicable while retaining an apparently benign and publicly defensible surface meaning.

References

• Clark, B. (2022). Pragmatics: The Basics. Routledge.
• Saul, J. (2018). Dogwhistles, political manipulation, and philosophy of language. In D. Fogal, D. W. Harris, & M. Moss (Eds.), New Work on Speech Acts (pp. 360–383). Oxford University Press.
• 山口智美, 斉藤正美, & 津田大介. (2023). 宗教右派とフェミニズム. 青弓社. [Yamaguchi et al. (2023). The Religious Right and Feminism. Seikyusha.]