The Social Coordinates of Korean Addressee Honorifics and Generational Conflict

The Social Coordinates of Korean Addressee Honorifics and Generational Conflict

Centered on the Egalitarian Function of the -yo Style and a 'Five-Axis Speech Model'

Abstract

This essay interprets Korean addressee honorifics not as a mere system of etiquette but as a sociolinguistic device that grammatically indexes the relationship between speaker and listener at the moment of each utterance. Through sentence-final endings, terms of address, particles, honorific pre-final endings, and lexical choices, a Korean speaker simultaneously adjusts the hierarchy, distance, and intimacy vis-a-vis the listener, the directionality of requests and directives, and an interactional stance. This paper conceptualizes this as a 'five-axis speech model' and uses the model to reinterpret generational conflict in Korean society. In particular, the -yo style, as the central mode of informal addressee honorification that developed within Korean, can function as an egalitarian device that maintains respect while easing excessive hierarchization. This paper's claim is not that Korean inevitably justifies hierarchy. Rather, precisely because Korean has so finely grammaticalized relationships, contemporary Korean society must more explicitly distinguish the domains in which hierarchy should operate from the domains in which equality must be guaranteed. From this perspective, generational conflict cannot simply be reduced to 'the authoritarianism of the older generation' or 'the rudeness of the younger generation'; it needs to be understood as a collision of speech defaults formed through the language socialization processes of different eras.

Keywords: Korean, addressee honorifics, the haeyo style, the hapsipsio style, generational conflict, sociolinguistics, hierarchy, equality, pragmatics, Korean AI

1. Framing the Problem: Korean Is a Language That Inscribes Relationships at the End of the Sentence

In Korean, an utterance does not end with the transmission of a proposition. The very same event of 'went' can be realized in multiple forms: gatda, gasseo, gasseoyo, gatseumnida, gasyeosseoyo, gasyeotseumnida. This difference is not a mere stylistic variation. Each form simultaneously reveals what kind of relationship the speaker places the listener in, how formal the speech situation is, and how much respect and distance is granted to the other party.

Korean addressee honorifics systematically mark the relationship with the listener at the sentence's closing. According to the National Institute of Korean Language, styles such as the hapsipsio, hao, hage, and haera styles belong to the formal register, while the haeyo and hae styles belong to the informal register. The formal register conveys a ceremonial, direct, assertive, and objective feel, whereas the informal register conveys a relatively soft and subjective one. Moreover, Korean addressee honorification is not a simple matter of lexical choice but a grammatical category realized through sentence-final expressions; and while the 'yo' of the haeyo style can, depending on one's mode of analysis, be regarded as an auxiliary particle, on the whole it functions as a device for realizing addressee honorification.

The purpose of this essay is to extend this grammatical fact into a social analysis. With every utterance, a Korean speaker positions the relationship with the listener on a certain set of coordinates. These coordinates are not simply a one-dimensional axis of 'raising/lowering.' Hierarchy, distance, intimacy, the directionality of requests and directives, and interactional stance all operate at once. This paper formalizes this composite judgment as a 'five-axis speech model' and uses it to reinterpret Korea's generational conflict from the standpoint of linguistic structure.

2. Historical Background: The Creation of a Script and a Hierarchical Society Coexisted

Korean's honorific system is not a device artificially designed at a particular point in time; it was formed through the long-term interaction of social hierarchy and linguistic convention. Hangul-that is, Hunminjeongeum-is known as a writing system created by King Sejong in 1443 and promulgated in 1446. UNESCO and materials related to Korean cultural heritage explain that Hunminjeongeum was completed in 1443 and promulgated in documentary form in 1446.

Yet the creation of a script did not immediately mean the egalitarianization of the entire speech community. Joseon was a society in which the status system, the bureaucracy, and the Confucian family order operated powerfully, and language use did not exist apart from that social order. Thus Hangul was a revolutionary invention that expanded ordinary people's access to writing, even as the society in which that script was used remained hierarchically organized. It is important to distinguish these two points. The accessibility of a writing system and the equality of speech relationships are not the same problem.

It is precisely here that Korean addressee honorifics take on important sociolinguistic significance. Korean addressee honorification repeatedly marks, at the end of the sentence, how the speaker treats the listener. In other words, in Korean a relationship does not remain background information external to the utterance; it enters into the structure of the sentence itself. Recent research likewise argues that addressee-honorific endings such as Korean's -seumnida and -yo can be seen as a grammatical category that demands a relational judgment before one even speaks.

3. Theoretical Framework: Power, Solidarity, and the Separation of Spheres

Analyzing Korean addressee honorifics requires two theoretical frameworks. The first is Brown and Gilman's 'power and solidarity' model. In analyzing the choice of second-person pronouns in European languages, they held that terms of address and reference are not mere grammatical choices but social indices of power relations and intimacy. Brown and Gilman's classic paper is summarized as holding that pronoun choice reflects relations of power and solidarity. In Korean, this function is realized over a far wider range than pronouns-particularly through the systems of sentence-final endings and terms of address.

The second is Michael Walzer's theory of the 'spheres of justice.' Walzer held that social goods must not be distributed according to a single criterion, and that each sphere-money, power, education, honor, family, politics, and so on-has its own distinct principle of distribution. His central concern was to block the 'convertibility' by which dominance in one sphere comes to govern all the others. This view is useful for analyzing the problem of hierarchy in Korean. Authority arising from one's job may exist, but it must not encroach upon the sphere of human dignity. Differences in expertise may be acknowledged, but they must not be converted into a right to demean someone as a person.

The core thesis of this paper, then, is as follows. Korean is a language capable of marking hierarchy grammatically, but for that very reason, modern society must all the more clearly separate the domains in which hierarchy operates from the domains in which equality must be guaranteed.

4. The Sociolinguistic Meaning of the -yo Style: The Egalitarian Potential of Informal Honorification

In Korean's honorific system, the -yo style occupies a special position. According to the National Institute of Korean Language's classification, the hapsipsio style belongs to the formal register and the haeyo style to the informal register. It is therefore not accurate to call the two systems 'exactly the same speech level.' Still, they have in common that both realize honorification toward the listener. The difference is less one of the presence or absence of respect than one of formality, situational register, and sense of distance.

For example, gamnida strongly carries formality, assertiveness, and an institutional sense of distance. Gayo, by contrast, makes the speech situation less ceremonial without lowering the listener. Ga can be intimate, but it removes addressee honorification. For this reason the -yo style functions as an important buffer in modern Korean. It is not a mere intermediate step lying between the intimacy of plain speech and the respectfulness of the hapsyo style, but a speech form that embodies respect and approachability at the same time.

Recent research on family discourse likewise points out that Korean honorific markers do not necessarily reflect only traditional hierarchy or etiquette, and that they can be used for a variety of discourse strategies-equality, face-saving, adjusting the speaker's stance, emotional expression, and so on. In family KakaoTalk conversations in particular, honorific markers function as devices not only for authority relations but also for relational adjustment and identity construction. This shows that viewing the -yo style simply as 'low-grade polite speech' is not sufficient.

The social meaning of the -yo style can therefore be summarized as follows. The -yo style is a core egalitarian resource of modern Korean: it respects the listener while reducing excessive institutional distance, and it creates intimacy while avoiding condescension.

5. The Five-Axis Model of Korean Speech

When a Korean speaker selects a sentence, what is being decided is not a single honorific grade. In actual speech, at least the following five axes operate simultaneously.

Axis Description Example question

1. Interactional stance Degree of seriousness, humor, joking, indirectness Is the present situation formal or casual?

2. Hierarchy/relationship Age, position, role, expertise, family relations Do I place the other person above, below, or as an equal?

3. Social distance Unfamiliarity, familiarity, public distance, private distance Will I keep distance from the other person, or close it?

4. Directionality of action Favor, request, directive, command, suggestion Who is demanding what of whom?

5. Emotional intimacy Liking, affection, burden, defensiveness, solidarity Will I open my heart, or close it?

To unpack the table: when a Korean speaker produces an utterance, the choice does not stop at a simple distinction between polite speech and plain speech. In the actual process of speaking, social judgments along at least the following five dimensions operate simultaneously.

1. Interactional stance

The speaker first decides with what stance to perform the utterance. This concerns whether the utterance is a serious explanation, a light joke, an indirect suggestion, or a direct demand. Even with identical propositional content, the social effect of a sentence changes depending on the stance the speaker takes.

2. Hierarchical relationship

The speaker judges the relationship between themselves and the listener. Age, position, role, expertise, family relations, and the like all bear on this judgment. In Korean, such relational judgments are directly reflected in the choice of sentence-final endings, terms of address, and honorific expressions.

3. Social distance

The speaker adjusts the distance from the listener. The manner of speech used toward someone met for the first time, someone in a public relationship, and an intimate differs in each case. Here the sense of distance is connected not only to mere closeness or remoteness but also to the formality of the speech situation, the institutional context, and psychological distance.

4. Directionality of action

The speaker sets the character of the action contained in the utterance. The form of the sentence changes according to whether an utterance is a favor, a request, a suggestion, a directive, or a command. In Korean, even the same demand for an action is expressed in various ways-haejwoyo, haejuseyo, haejusigesseumnikka, haseyo-and each expression implies a different power relation and degree of imposition.

5. Emotional intimacy

The speaker modulates the emotional distance from the listener. An utterance carries emotional elements such as liking, affection, burden, defensiveness, and a sense of solidarity. Thus some expressions are polite yet create emotional distance, while others are less formal yet form a stronger sense of intimacy and solidarity.

In this way, speech choice in Korean is not a single grammatical choice but a process of setting composite social coordinates. With every sentence the speaker simultaneously adjusts interactional stance, hierarchical relationship, social distance, directionality of action, and emotional intimacy. Korean addressee honorifics, therefore, need to be understood not as a mere system of etiquette but as a grammatical and pragmatic device that organizes, in real time, the social relationships within a speech situation.

For instance, the utterance igeo jom haejwo signals high intimacy and close social distance, and in some cases can give the impression that the speaker is issuing an informal directive to the listener. Igeo jom haejwoyo maintains intimacy while inserting addressee honorification. Igeo jom haejuseyo lowers intimacy and raises the politeness of the request. Igeo haejusimyeon gamsahagesseumnida greatly raises both social distance and formality. Igeo hae may be natural in an intimate relationship, but in a hierarchical situation it can be interpreted as a command.

The important point is that all of these variants can convey identical propositional content. Their social meanings, however, are completely different. This is precisely where the difficulty of Korean lies. Knowing the vocabulary and grammar is not enough. Each time, the speaker must compute the social coordinates of the utterance.

This model also has direct implications for Korean-language education and for Korean AI. Translating and generating Korean honorific expressions requires information about the speaker-listener relationship, the discourse context, and surrounding sentences. Indeed, research on the machine translation of Korean honorific expressions proposes that one must draw on surrounding context and speaker-relationship information in order to handle honorific expressions more appropriately. Research on the evaluation of Korean LLMs likewise raises the problem that translation-centered benchmarks alone make it difficult to adequately capture Korean's morphosyntactic and pragmatic characteristics.

6. Reinterpreting Generational Conflict: A Collision of Speech Defaults, Not of Character

We can now apply this model to generational conflict in Korean society. Korea's generational conflict is often expressed in moral language. The younger generation criticizes the older generation as 'authoritarian,' while the older generation judges the younger generation to be 'rude.' But such accounts have the limitation of reducing the conflict to a matter of individual character.

From this paper's perspective, generational conflict can in large part be understood as the result of differing language socialization. Even when they use the same Korean, different generations may have different default speech coordinates. A generation socialized in an era when industrialization, military culture, corporal punishment in schools, and the patriarchal family order were powerful may tend to activate the axes of hierarchy and distance more strongly. By contrast, a generation socialized after democratization-amid digital communication, horizontal organizational cultures, and the discourse of individual rights-may use the axes of intimacy and equality more naturally.

Such generational differences are also partly confirmed in actual surveys of language use. According to a 2024 survey by the National Institute of Korean Language-covering 3,000 men and women aged 15 to 69 nationwide on terms of address and everyday expressions-differences in expressive preferences by age emerged clearly. For example, the older the age group, the more they preferred 'agassi' (miss) when addressing a young female sales clerk, whereas younger people tended to prefer expressions that mark gender and age less directly, such as 'yeogiyo/jeogiyo' (excuse me) or 'sajangnim' (boss/sir).

This data suggests that generational differences in linguistic sensibility are not a mere matter of taste but a difference in how social relationships are encoded. To one generation a particular term of address may be a natural courtesy, while to another that same expression may sound like a gendered, age-marked, and hierarchizing form of hailing. Conflict arises at exactly this point. One side feels that they 'just spoke naturally,' while the other feels, 'why are you placing me in such a position?'

What is needed to reduce generational conflict, then, is not moral condemnation of one side or the other. What is needed is a metalanguage that recognizes differences in speech coordinates. Questions like the following must become possible: 'Am I right now activating the hierarchy axis too strongly?'; 'Can the other person receive my -yo style not as rudeness but as informal honorification?'; 'Is this situation the domain of job hierarchy, or the domain of human dignity?'

7. The Principle of Separating Spheres: Job Hierarchy and Human Dignity Are Different

The point at which Korean's honorific system causes problems in modern society is not the fact that hierarchy exists at all. The problem arises when the hierarchy of one sphere is transferred into another.

A company department head may hold the authority to issue work directives. A doctor may hold explanatory authority grounded in specialized knowledge. A teacher may hold authority over the running of a class. But the moment this authority turns into a right to lower the other person's human dignity, it becomes a problem. Job-based authority belongs to the sphere of roles, while human dignity belongs to the sphere of fundamental rights. The two spheres must be separated.

In this respect, the issue of plain speech versus polite speech in Korean is not a mere matter of speech style. In an organization where a superior continually uses plain speech toward a subordinate, job authority easily seeps into the sphere of personhood. In a family where parents use only one-sided plain speech toward their adult children for life, it is hard for the relationship of protection and nurturing to transition into one of mutual respect between adults. Conversely, the attitude of regarding every hierarchical expression as violence also creates problems-because in the workplace there are times when differences in responsibility, authority, and expertise must be clearly expressed.

The solution, therefore, is closer to 'let us limit the domains in which hierarchy operates' than to 'let us abolish hierarchy.' In the domains of job directives, professional judgment, and the allocation of responsibility, a certain degree of hierarchical expression can be functional. But in the domains of personal evaluation, basic dignity, the right to speak, safety, and legal rights, the principle of equality must take precedence. This is the result of applying Walzer's principle of separating spheres to the ethics of Korean speech.

8. Practical Implications: A -yo-Style Mode and Education in Speech Coordinates

This paper's analysis yields several practical proposals.

First, it is possible to experiment with making the -yo style the default in organizational communication. The -yo style is less rigid than the hapsipsio style and less hierarchical than the hae style. In situations calling for horizontal collaboration, then, the -yo style can serve as a linguistic device that raises psychological safety within an organization. That said, in situations requiring institutional clarity-official reports, legal documents, external presentations, and the like-the hapsipsio style remains appropriate.

Second, intergenerational communication education should be education in speech coordinates, not education in etiquette. The approach of 'speak this way to those above you' and 'you may speak this way to those below you' merely repeats the problem. Instead, learners should be made to analyze: 'How strongly is the present utterance activating the hierarchy axis?'; 'Will the other person receive these words as intimacy or as condescension?'; 'Is this request a favor or a directive?'

Third, Korean AI systems must move beyond being mere polite-speech converters to become inferers of relational coordinates. For Korean AI to operate naturally, superficially converting among the haeyo, hapsipsio, hae, and hage styles is not enough. It must also infer the speaker-listener relationship, the formality of the situation, the imposition of a request, emotional distance, and the context of the prior conversation. Recent research on generative AI and Korean honorifics has likewise begun attempting to compare differences among the inputs of Korean honorific expressions, corpora, learner assessments, and AI-generated sentences.

9. Limitations and Directions for Future Research

This paper is a theoretical essay and not yet a completed empirical study. The following follow-up research is therefore needed.

First, generational differences in speech coordinates should be verified on a corpus basis. One should collect actual meeting conversations, family conversations, online conversations, and customer-service conversations from speakers in their 20s, 30s, 40s, and 50s and older, and compare the distributions of sentence-final endings, terms of address, request forms, and command forms.

Second, the egalitarian function of the -yo style should be verified experimentally. One can measure, by generation, how the listener's sense of respect, distance, authority, and imposition changes when the same request is presented in the hae, haeyo, and hapsipsio styles.

Third, Korean AI evaluation metrics should include the accuracy of relational coordinates. Many current Korean AI evaluations focus on semantic accuracy, grammaticality, general knowledge, and translation quality. But what actually matters to Korean users are questions like 'the words are correct, but is it off-putting?', 'it is polite, but does it feel too distant?', and 'it is friendly, but does it cross a line?' These should be treated as an independent axis in the quality evaluation of Korean AI.

10. Conclusion

Korean is a language that marks relationships grammatically. Addressee honorifics in particular reveal, with every utterance, the social position in which the speaker places the listener. This structure is at once the richness and the burden of Korean. With every sentence, a Korean speaker must simultaneously adjust hierarchy, distance, intimacy, directionality of action, and interactional stance.

This essay formalized that structure as a 'five-axis speech model' and applied it to an analysis of generational conflict in Korean society. Generational conflict does not arise simply because someone's character is bad. In large part it arises because speakers socialized in different eras use different speech defaults. The hierarchical manner of speech of the older generation is often the result of the natural language they learned, and the -yo-centered speech of the younger generation is likewise the result of the egalitarian linguistic environment they learned.

But this explanation grants no one a free pass. The fact that a language was historically formed does not mean that all of its linguistic habits are justified. On the contrary, when we understand the process of historical formation, we can adjust it more consciously. When we separate job hierarchy from human dignity, harness the egalitarian potential of the -yo style, teach the five-axis coordinates of speech, and understand intergenerational linguistic differences as structural differences rather than as grounds for moral condemnation, the communication costs of Korean society can be reduced.

Korean is a language that bears deep traces of hierarchy. Yet at the same time, within it there also exist resources for speaking of equality. The question is not whether Korean is hierarchical or not. The question is how consciously we are able to use that language.

References (Draft)

Brown, R., & Gilman, A. (1960). The pronouns of power and solidarity. In T. A. Sebeok (Ed.), Style in Language (pp. 253-276). MIT Press.

Brown, P., & Levinson, S. C. (1987). Politeness: Some Universals in Language Usage. Cambridge University Press.

Lee, Iksop, & Ramsey, S. R. (2000). The Korean Language. SUNY Press.

Sohn, Ho-min. (1999). The Korean Language. Cambridge University Press.

Walzer, M. (1983). Spheres of Justice: A Defense of Pluralism and Equality. Basic Books.

National Institute of Korean Language. 'Addressee Honorification and the Auxiliary Particle Yo,' Online Gana-da, 2024.

National Institute of Korean Language. 'The Haera Style of the Formal Register,' Online Gana-da, 2025.

National Institute of Korean Language. 'What Words Do Koreans Use in These Situations? Vocabulary Varies by Age and Gender,' 2024.

Kim, J. (2022). Corpus-based analysis of honorifics in Korean and its pedagogical implication. Korean Linguistics.

Kim, J., et al. (2021). Context-aware neural machine translation for Korean honorific expressions. Electronics, 10(13), 1589.

Park, C., et al. (2025). Open Ko-LLM Leaderboard2: Bridging foundational and practical evaluation for Korean LLMs.