Rinelle Chen

This paper attempts to reread selected passages from Jorge Luis Borges’s The Library of Babel and The Garden of Forking Paths, together with Italo Calvino’s Invisible Cities, from the perspective of the global–local structure. Its aim is not to offer a total interpretation of the two writers’ works, but to propose a limited comparative path: to observe how conceptual form, global structure, and human desire generate different local destinies. This paper argues that Borges’s two texts present a form of local collapse within a perfect concept. In The Library of Babel, the library, as a global textual system, preserves all possible books; at the same time, it also preserves truth, error, catalogues, false catalogues, and refutations. The problem is not the absence of meaning, but the fact that truth and noise are both incorporated into the same complete structure, making it impossible for the local subject to maintain a stable capacity for discernment. In The Garden of Forking Paths, Ts’ui Pên’s labyrinth of time preserves all possible paths; however, this global model of time does not exclude the irreversibility with which Yu Tsun completes an action, causes a death, and bears guilt along this particular path. From this perspective, the global structures in Borges’s texts often approach conceptual perfection. Yet when human desire attempts to find within them a position from which something can be possessed, proven, or completed, the local instead moves toward collapse and destruction. By contrast, the relevant passages in Calvino’s Invisible Cities present a form of local manifestation within an unguaranteed concept. The net of Octavia does not eliminate the abyss, yet it forms a limited support. Leisa is not a happy city, yet within unhappiness it presents brief lines of life-connection. The carpet of Eudoxia does not obtain final authority, yet it allows the city and the schema to remain in an unresolved relation of mutual illumination. Calvino does not use a perfect globality to guarantee local meaning; rather, he allows the local to manifest temporarily under conditions of incompleteness, instability, and the absence of final guarantee. Human desire here does not need to complete the totality, yet it can still show its human dimension through viewing, dwelling, connection, and support. Through this comparison, this paper suggests that the difference between Borges and Calvino lies not only in whether the global is complete, but also in the relation between conceptual form and human desire. Borges presents the severe pressure exerted by perfect concepts upon local desire; Calvino, by contrast, presents the possible manifestation of humanity within unguaranteed forms. Meaning is therefore not only a question of whether it exists, but of how local life enters global structures, endures them, collapses under them, or re-manifests itself within them.

This paper takes Kafka’s “Before the Law” and the film The Reader as its objects of analysis, and proposes threshold dependency as a structural framework for understanding the unfinished relations in these two works. On the surface, “Before the Law” concerns law, the door, the doorkeeper, and waiting; The Reader concerns postwar guilt, intimacy, reading aloud, education, and historical responsibility. This paper argues, however, that beneath their differences at the narrative surface, the two works present a shared relational structure: the relation can neither be completed by crossing the threshold nor terminated by the threshold’s closure. Instead, the threshold itself becomes the minimal condition that allows the relation to continue existing. What this paper calls threshold dependency is not an ordinary relation of obstruction. An ordinary threshold presupposes that crossing is possible: once the subject satisfies the required conditions, they may enter the next stage. A closed threshold, by contrast, presupposes that passage has been blocked, and that the relation has been excluded or terminated. In threshold dependency, however, the threshold can neither be crossed nor closed. It prevents the relation from being completed, while also preventing the relation from disappearing. The relation is therefore fixed in a form of unfinished proximity: the subject can orient themselves toward the object, but cannot truly enter; they can maintain a connection, but cannot transform it into a stable, explicit, or settled relational form. In Kafka’s “Before the Law,” this structure appears as follows: the door is open, but passage does not take place. The man from the country does not face a completely closed door, but rather an entrance that is visible, available as an object of orientation, and even “meant only for him.” Yet precisely because the door remains open and impassable, he can neither understand himself as simply excluded nor complete the act of entering the law. His life is therefore fixed before the door: waiting, requesting, bribing, gazing, and aging do not lead to true entry, but instead maintain the threshold relation between him and the law. The door here is not an obstacle outside the law, but the only relational condition through which the man from the country and the law can continue to be connected. In The Reader, threshold dependency appears in a different form. The relation between Michael and Hanna is not completely severed, yet it also cannot be transformed into ordinary intimacy, reconciliation, or shared life. Reading aloud, recording, and distance constitute a mediated threshold: Michael’s voice can reach Hanna, but his body, shared life, and an explicit naming of the relation cannot; life can be arranged, but reception cannot take place; connection can be maintained, but the relation cannot be completed. This relation depends on the threshold formed by voice and distance. Once Hanna is released from prison and this threshold of distance can no longer be maintained, the relation must face the pressure of becoming real, ethical, and nameable; yet this pressure is precisely what the original form of the relation cannot bear. The relation between Michael and Hanna cannot take the form of an ordinary connection, because it is torn open by multiple asymmetries of age, knowledge, shame, guilt, time, and perspective. If they were to be together again, the relation would be forced into a real ethical framework, which would instead cause it to collapse. Through these two cases, this paper attempts to show that certain relations in literary and cinematic narratives do not take completion, reconciliation, arrival, or rupture as their stable form. Rather, they depend on a threshold state in order to maintain their existence. In Kafka, the door keeps the man from the country in the position of being “before the law”; in The Reader, voice and recording keep Michael and Hanna in a position of being “related yet unnameable.” The former cannot say, “I have entered the law,” or “I have been completely excluded.” The latter also cannot be reduced to “I love her, and therefore forgive her,” or “She is guilty, and therefore I cut her off.” What both works present is a relational structure that cannot be completed, cannot be settled, and cannot truly be left behind. This paper therefore argues that threshold dependency offers a new entry point for understanding unfinished relations: the threshold is not an obstacle outside the relation, but a sustaining condition within the relation itself. Some doors are not there to be passed through, but to make one remain; some voices are not there to complete an encounter, but to allow a relation to continue under conditions in which completion is impossible.

This paper examines like, a highly frequent English expression with a complex functional distribution, and proposes a minimal structural analysis from the perspective of understanding formation. Existing accounts often classify like into functional categories such as similarity, approximation, exemplification, quotation, discourse marking, or preference expression. These classifications are descriptively useful. However, if the analysis remains only at the level of functional distribution, it may obscure the local mode of relation formation shared across different uses. Based on a small-scale observation of natural language data, this paper proposes that the core role of like should not be understood merely as “similarity” or “preference.” Rather, it can be described as a form of unspecified shared-feature bridging. In approximate and discourse-related uses, like connects a target with subsequent content, allowing certain shared features in the subsequent item to enter the understanding of the target. Crucially, like does not specify which particular feature-line has been activated. As a result, the subsequent item does not enter understanding as a flat literal value, but remains accessible through multiple possible facets. This paper describes this pattern as a form of prismatic shared-feature bridging: the multi-faceted accessibility produced by like is not its primary function, but a structural effect of unspecified bridging. This paper further argues that although verbal like grammatically fixes the relation as one of liking or disliking, the features supporting that orientation remain unspecified. For example, I like him or I don’t like your answer does not specify whether the orientation arises from appearance, personality, tone, logic, attitude, or some other feature. What these expressions fix is a positive or negative orientation, not the specific feature-line supporting that orientation. Thus, although approximate like and verbal like occupy different structural positions, they still share a deeper local pattern: a relation is established, while the shared feature-line supporting that relation is not locked into a single path. This paper does not attempt to establish a complete lexical theory of like, nor does it claim that all uses of like can be reduced to a single model. Instead, it takes uses such as approximation, quotation, perceptual judgment, manner-of-treatment expressions, preference, and numerical approximation in natural language data as points of observation, and proposes a structural hypothesis that can be further tested: in understanding formation, like often establishes a relation through unspecified shared features, thereby allowing the subsequent item, object, or source of orientation to remain in a state of multi-faceted accessibility. Through this analysis, the paper aims to show that small linguistic expressions not only mark semantic or pragmatic functions, but may also participate in the local organization of relation formation, feature emergence, and multi-path accessibility in understanding.

This paper examines the structural operation of the common English expression already in understanding formation. Existing analyses often treat already as a marker of completion, prior occurrence, state establishment, or earlier-than-expected timing. Such descriptions have considerable explanatory value and can account for the basic effects found in sentences such as I already told you, She already knows, or Sunday morning already? However, when already is further examined in commands, urging contexts, the early emergence of affective positions, and cases involving “having already come this far” or “no longer being able to delay,” completion or earlier-than-expected timing alone is insufficient to explain how it actually operates in understanding. This paper argues that the core of already should not be understood merely as completion marking, but can be described as a crossed-side positioning operation. By crossed-side positioning, I refer to the operation by which already makes the relevant content no longer understood as still remaining on the pre-crossing side, but instead requires it to be repositioned on the other side of a boundary. In this process, the pre-crossing side is no longer understood as a position in which understanding can naturally remain. This “crossed-side” status is not equivalent to grammatical perfect aspect, nor does it necessarily mean that an event has been objectively completed. Rather, it means that the position of understanding has been required to stand on the other side. Through a small-scale observation of natural-language data, this paper distinguishes several local forms of already. First, in its basic uses, already makes the side of the unknown, the untold, the unfamiliar, or the not-yet-entered no longer naturally available, and positions understanding on the side of the known, the told, the familiar, or the entered. Second, its uses in temporal, life-stage, and affective positions show that already can mark a position as having to be acknowledged as located on the other side, even when subjective expectation has not yet adjusted to it. Third, interactional urging uses, such as Say something already or Shut up already, clearly show that already does not describe an event as completed, but requires the other party to leave the original side and cross into the side where something should hold or should stop. Fourth, uses involving excessive load, advanced procedural positioning, or opportunities that can no longer be retained show that already can mark a load, institutional process, commitment, or opportunity path as having crossed a critical point, such that it can no longer be handled from the original side. This paper does not aim to establish a complete lexical theory of already, nor does it attempt to exhaust all of its grammatical and pragmatic distributions. Its goal is more limited: through a set of observable examples and a minimal structural description, it aims to show that already repeatedly exhibits a stable local operational core across different contexts. This core is not lexical monosemy, but stability at the operational level: already makes the pre-crossing side no longer naturally available and repositions understanding on the other side. Through this analysis, the paper seeks to show how small linguistic expressions participate in the reconfiguration of positions of understanding, and to provide an initial structural description of local crossing and repositioning in understanding formation.

This paper examines two highly frequent English expressions, almost and too, whose modes of operation in understanding formation differ. Existing studies often treat almost as a marker of near completion, approximation, or incomplete attainment, and treat too as a degree expression indicating excess, overextension, or exceeding a standard. However, such analyses often remain at the level of semantic content or degree function, and less often address how these expressions form local structures of progression in the process of understanding. Based on a small-scale observation of natural language data, this paper proposes a minimal structural distinction: almost may be understood as the formation of boundary pressure, whose function is to bring a state close to a threshold, point of completion, or categorical boundary, while still keeping it on the side of non-completion. Its core does not lie in completion itself, but in the state of tension formed near the boundary: the relation has moved very close to the point of crossing, but has not actually crossed. In some contexts, this incomplete state of approach may still leave downstream effects, residual pressure, or eventive traces. By contrast, too may be understood as a structural operation of disequilibrium induction. Its function is not merely to raise degree, but to mark a property, intensity, or state as exceeding the accommodative proportion of the context, thereby producing instability, resistance, blockage, difficulty of accommodation, or a shift in downstream understanding. In structures such as too … to V, this disequilibrium is further connected to an action path, such that an originally continuable direction of action becomes obstructed or suspended. This paper therefore argues that the difference between almost and too should not be understood merely as a difference in degree level or semantic function. Rather, it involves two distinct forms of local structural dynamics: the former develops around boundary pressure of approaching without crossing, while the latter develops around a state of disequilibrium formed after exceeding accommodative proportion. Through this minimal contrast, the paper further suggests that small linguistic expressions do not only participate in semantic modification. They may also participate, within local progression in understanding, in the formation of threshold approach, proportional disequilibrium, and tension distribution.

This paper examines two highly frequent English expressions that participate differently in understanding formation: if and maybe. Existing studies often treat if as a conditional connective and maybe as a marker of uncertainty or modality. However, such analyses largely remain at the level of semantic content or pragmatic function, and less frequently address how these expressions influence the formation and stabilization of relations during understanding. Based on small-scale observations of naturalistic data, the present paper proposes a minimal distinction. If may be understood as the introduction of a process of conditional structuring, whose role is to allow subsequent relations to unfold along a particular condition. By contrast, maybe maintains a proposition in a state of incomplete settlement, allowing related content to be introduced and considered without converging into a fixed judgment. Furthermore, the paper argues that the differences associated with if do not primarily arise from functional categories themselves, but from whether the introduced conditional structure further develops into a continuable pathway of understanding. In some cases, the condition may externally unfold into a more explicit continuation structure; in others, it remains unexpanded, such as when embedded within cognitive expressions like I wonder or I don’t know. On this basis, the paper suggests that differences between linguistic expressions are not limited to semantic content or pragmatic function, but also involve how they organize conditionality and incomplete settlement within understanding formation. Through a minimal contrast between if and maybe, the paper proposes a preliminary framework in which linguistic understanding may be described as a dynamic organization between conditional structuring and incomplete settlement, rather than merely as a collection of semantic markers or functional categories. The paper does not attempt to establish a complete lexical theory, nor does it claim uniqueness for the proposed schematizations. Its aim is more limited: to show, through observable minimal contrasts, that small linguistic expressions may continuously participate in the local organization of understanding.

This paper examines two highly frequent English expressions—actually and anyway—which exhibit subtle but systematic differences in how they operate in the formation of understanding. While existing analyses typically classify them as emphasis markers, discourse markers, or pragmatic devices, such categorizations may obscure their role in shaping the dynamics of interpretation. Based on observations from a small set of naturally occurring data, this paper proposes that actually and anyway do not merely mark different semantic or pragmatic functions, but instead participate in distinct local operations within an ongoing interpretive process. Actually can be described as a form of epistemic re-alignment, introducing a momentary adjustment to an existing interpretive trajectory, after which the trajectory proceeds without further structural obstruction. In contrast, anyway can be described as a form of constraint de-binding, through which previously relevant conditions lose their constraining force, allowing the trajectory to detach from prior constraints and continue independently of them. This paper does not aim to construct a comprehensive lexical theory of actually and anyway. Rather, through a set of minimal contrasts, it shows that even when expressions appear functionally similar at the level of discourse or pragmatics, they may still differ in how they participate in the formation of understanding. On this basis, the paper argues that small linguistic units can function as local operators in interpretive dynamics, continuously shaping alignment and restructuring constraint relations at a micro level.

This paper proposes a conceptual framework for analyzing how responses unfold along structured pathways of understanding, rather than being treated as associations between discrete nodes or as one-time retrieval events. The aim is not to report empirical findings, but to introduce a set of structural hypotheses that can be observed and tested. Within this framework, the paper introduces pathway accessibility as a central concept, referring to the condition under which multiple potential pathways are differentially accessible in a given moment of understanding. Pathway accessibility does not correspond to selection probability. Instead, it describes the competitive distribution among pathways and their relative availability for continuation. The pathways proposed here are not exclusive structures. Rather, they form a generative descriptive layer that coexists with existing relational structures. Accordingly, understanding is not characterized as transitions between discrete nodes, but as a process that unfolds along multiple accessible pathways, exhibiting path-dependent generation. Furthermore, shared properties are treated as conditions that may exist within relational structures, but their manifestation in understanding is neither uniform nor automatic. Instead, they become foregrounded through the activation and continuation of specific pathways. In this sense, understanding does not operate by extracting pre-existing similarities, but by selectively amplifying and rendering visible certain relational axes under given conditions. When disruption or blockage occurs, it should not be interpreted as a simple failure of retrieval, but rather as a case of field reconfiguration, which alters the subsequent distribution of pathway accessibility.

This paper examines three English expressions associated with wish and non-current states—if only, I wish, and the noun wish—and suggests that their differences should not be understood merely in terms of semantic content or emotional intensity, but rather in terms of how wishes are formed in understanding through different structural modes. In existing analyses, these expressions are often grouped under categories such as wish, counterfactuality, regret, or preference. While such classifications provide descriptive value, they tend to remain at the level of results and do not sufficiently address how these expressions form non-current states in understanding. This paper therefore proposes a structural distinction: if only, I wish, and wish (as a noun) should not be treated as different expressions of the same wish, but as corresponding to three distinct modes of formation. The paper first argues that if only should not be understood merely as the expression of a wish, but as a rebound event. In this structure, the utterance opens a non-current or unmet condition, while this condition is simultaneously pulled back by the reality baseline, producing a rebound tension. This rebound may vary in amplitude, ranging from strong regret and self-collapse to minimal openings (e.g., if only for a moment), suggesting that rebound is a structural feature rather than a by-product of high emotional contexts. By contrast, I wish does not primarily form a rebound, but establishes a projection link that points from the current position toward a non-current state. This projection should not be understood as a detached or purely imaginative construct; rather, it can be understood as a state formed through the distortion, reconfiguration, or preferential shaping of reality. In this structure, reality does not disappear, but remains present as the source of distortion, allowing the wish to persist as a relation that can be maintained and held in view. While this structure may appear in interpersonal, institutional, or immersive contexts, its structural core remains stable. Furthermore, the paper suggests that when wish appears as a noun, the fixation of a projection into a node within the field of understanding takes place. At this point, wish no longer maintains an ongoing projection path, but instead functions as a stable reference point in the field of understanding, and thus becomes available for articulation, guidance, conflict, or negotiation. This transformation is described as node fixation, and is suggested to play an important role in relational formation and social interaction. This paper does not attempt to establish a full typology of wish constructions, nor does it aim at statistical generalization. Rather, through small-scale observations of natural data, it proposes a minimal contrast: similar wish-related meanings may be formed through different pathways in understanding. In doing so, the paper shifts the analytical focus from what a wish is to how a wish is formed in understanding, and proposes a continuous model from rebound events, to projection links, to node fixation structures.

This paper examines two highly frequent English words that do not operate identically in the formation of understanding: even and yet. Existing analyses typically classify even under scalarity, focus, emphasis, or unexpectedness, and yet under temporality, contrast, concession, or incompletion. While these classifications are descriptively useful, they tend to obscure how these words actually guide the formation of understanding within sentences when treated solely as functional labels. The central observation of this paper is that even and yet do not merely express different semantic functions, but organize distinct local force relations within an utterance. Even typically appears at positions where a threshold or expected constraint remains valid—that is, where some form of resistance has not been removed. Its function is not to cancel this resistance, but to mark that a force nevertheless passes through it. Thus, in expressions such as “even for entry-level positions” or “even after you shower,” even does not simply intensify meaning; rather, it allows a threshold that should impose a limit to remain structurally active while still being crossed. By contrast, yet typically appears at positions where an expected relation has not reached its expected completion. In sentence-final uses, as in “no candidate selected yet,” yet maintains a projected endpoint in a state of non-release. In clause-internal uses, it often produces delay, deflection, or asymmetric linkage under obstruction. In other words, yet does not merely mark contrast or reversal; it signals that a relation which should have continued, completed, or converged encounters resistance in its progression, resulting in delayed or structurally misaligned outcomes. This paper does not aim to construct a comprehensive lexical theory of even or yet, nor does it claim that either word has a single invariant function across all contexts. Instead, it focuses on a set of recurrent local differences in everyday usage: even tends to mark penetration under still-active resistance, while yet tends to mark an expected relation that remains visible but obstructed. Through this distinction, the paper suggests that beneath familiar categories such as scalarity, emphasis, contrast, concession, and temporality, there exist more fine-grained mechanisms of how understanding is formed. The difference, therefore, does not lie in whether resistance is present, but in how resistance relates to the unfolding of the utterance. Even marks resistance as still active yet traversed; yet marks a projected relation whose completion remains blocked under conditions of resistance. Accordingly, even is provisionally described as a form of penetrative continuity, while yet is described as marking an expected relation that remains obstructed while maintaining pressure. The relation carried by the utterance has reached a condition under which release or overflow would be expected, yet its actual manifestation may take the form of delayed non-release (sentence-final yet) or locally obstructed overflow (clause-internal structures). The notion of “overflow” here does not refer to any physical process, but to a tension state that remains present and observable under conditions of obstruction. These descriptions do not reduce lexical items to physical models. Rather, they function as an observational vocabulary for describing how small words reshape thresholds, resistance, continuation, and closure in the formation of understanding within sentences.

This paper examines two highly frequent yet relatively undercompared expressions in everyday English: just and still. Existing research has often analyzed just as a focus marker, minimizer, hedge, or scalar restriction marker, while still has commonly been examined in relation to temporal persistence, continuation, or concessive marking. These approaches offer important insights into their semantic and pragmatic functions, but they have paid comparatively less attention to how the two expressions may organize different interpretive pathways during meaning formation. Drawing on a small-scale observational dataset from naturally occurring Reddit discourse (20 just cases and 20 still cases), this paper proposes a preliminary distinction. In emotionally charged and self-disclosive contexts, just often compresses multiple private reasons, emotions, experiences, and internal lines of reasoning into a more unified outward expression, while leaving much of its intermediate interpretive path less explicitly articulated. By contrast, still more often marks the continued presence of a state that would normally be expected to fade following the passage of time, relational closure, shifting social expectations, or changing external conditions, thereby making that lingering state more readily perceptible to others. To describe this contrast, the paper introduces two provisional structural models: prism-like compression for just and channel persistence for still. Rather than proposing a comprehensive lexical theory, this paper aims to show that superficially similar emotional expressions may emerge through different interpretive paths, and that seemingly minor lexical items may systematically participate in the compression, visibility, and continuation of interpretive processes.

This paper examines a structural phenomenon in questioning that has received relatively limited independent attention: some question sequences do not follow typical patterns of linear discourse progression. In many existing approaches, questions are primarily understood as information-seeking acts or interactional moves that unfold through sequential response structures. Within such frameworks, later utterances are often assumed to depend primarily on immediately preceding ones, thereby forming local linear chains. This paper identifies a different structural pattern. In some question sequences, later questions do not primarily respond to the immediately prior question. Instead, they repeatedly orient toward a problem center that remains not yet stably accessible and difficult to enter directly. Under such conditions, the relationship between questions is better described as center dependency rather than linear dependency. When this occurs, questioning may display non-linear exploratory behaviors rather than straightforward sequential progression. This paper therefore proposes that some questions can be understood as local topology events embedded within broader discourse: they temporarily alter local patterns of progression without determining the structure of the entire discourse. Building on this distinction, the paper differentiates three question states: open-field questions (the problem space remains underdetermined), constrained-path questions (the answer remains unknown but available pathways are relatively limited), and structurally closed questions (surface interrogatives in which the space of inquiry is already highly restricted). Methodologically, the paper draws on approximately ninety publicly available Reddit examples as materials for structured conceptual observation. The goal is not statistical generalization, but to examine whether these structural distinctions can be observed consistently across different contexts. This paper argues that not all questions open information gaps through linear progression. Some questions temporarily reorganize local discourse movement, and center dependency may represent an important but underexamined structural form of questioning.

This paper examines whether everyday English expressions that appear to perform similar pragmatic functions may nevertheless differ in the internal dynamics through which closure is formed. Focusing on “never mind,” “whatever,” and “it’s fine,” with “fine” introduced as a key test case, the paper analyzes expressions that commonly serve to end a topic, reduce conflict, withdraw a prior move, or stabilize an interaction. Although these expressions are often grouped under similar pragmatic functions, this paper argues that their differences should not be understood merely in terms of tone, politeness, emotional attitude, or degree of intensity. Rather, they involve distinct pathways through which closure is formed. Methodologically, the paper begins from phenomenological observation and then develops structural analysis. The tension diagrams used in this paper originate from the author’s phenomenological observation of tension shifts during the interpretation of utterances. They are not proposed as fixed geometric properties of expressions, nor as uniquely correct final forms. Instead, they function as a structural display method for observing differences that are often compressed under similar pragmatic labels. On this basis, the paper argues that “never mind” is not simply a form of ending, but more closely resembles an inward withdrawal after an unrealized unfolding; “whatever” is not merely indifference, but a decentering and outward dismissal of the interactional center; and “it’s fine” involves a regulatory process in which local fluctuation is followed by restabilization. The case of “fine” further shows that a single lexical item does not correspond to one fixed tension diagram: under different contextual loads, it may display inward withdrawal, outward dismissal, or stabilizing fluctuation. The paper therefore argues that tension diagrams should not be understood as fixed word-to-shape mappings, but as descriptions of local utterance-in-context dynamics. Its aim is not to establish a complete typology of English closure expressions, nor to claim universal uniqueness for the diagrams. Rather, through a minimal contrast set, it shows that even when expressions appear pragmatically similar, they may form closure through different structural pathways. In doing so, the paper offers a structural entry point for describing everyday interactional expressions beyond semantic function labels and emotional categorization.

本文探討表面上可被概括為相近接受語義的中文語句，是否仍可能在理解中呈現不同的內部張力結構。本文以「也不是不行」、「也還可以」與「勉強可以」三句作為核心分析材料，並引入「可以接受」作為對照組。這四句在日常理解中皆可落入接受語義區域，因此容易被壓縮為相近表達；然而，本文主張，它們之間的差異不宜僅被理解為接受程度的高低，亦不只是語氣或風格差異，而更涉及可接受性如何被形成的結構差異。 為凸顯此一差異，本文採取體感先行、結構分析隨後跟進的寫作方式。文章一開始先呈現四句語句與對應的張力圖式，使讀者先經歷相近語義與異構圖式之間的落差，再進一步提出概念性描述。本文所使用的張力圖式，來自作者對語句理解過程中張力變化的體感觀察，並作為一種現象性的描述工具，用以呈現不同語句在可接受性形成上的結構差異。本文不將其宣稱為語句的幾何本體，亦不主張其構成唯一正確的最終形式；它們較適合被理解為一種現象性的結構顯示方式，用以穩定呈現不同語句在理解中所顯出的張力差異、分岔程度、阻力型態與收斂方式。依此觀察，本文指出：「也不是不行」呈現可能性展開後的震盪收斂；「也還可以」呈現快速局部開放後的穩定收斂；「勉強可以」呈現高阻力拖行後的低位通過；相對地，「可以接受」則呈現門檻確認的平滑整合。 本文的目的不在於建立完整的接受句類型學，也不在於主張圖式具有唯一正確性，而在於透過一組最小對比指出：即使語義表面高度接近，語句仍可能經由不同的張力路徑形成其可接受性。藉由此一初步分析，本文嘗試指出：表面近義並不必然意味相近的張力形成方式，並為中文接受句提供一個不同於強度刻度模型的結構性描述入口。

This paper proposes a punctuation analysis framework centered on the notion of sentence fields and the flow of tension, offering a re-examination of the roles of the period, ellipsis, and em dash at sentence-final positions. Rather than treating punctuation as an addition of tone, a cue for pause, or a purely conventional writing device, this paper argues that punctuation actively participates in the formation of boundary conditions within a sentence field, and systematically alters how tension flows within that field. From this perspective, the differences among the period, ellipsis, and em dash should not be described merely in terms of “whether a sentence ends” or “degrees of emphasis.” Instead, they are better understood as three distinct topologies of tension flow. The period produces a sustained closed-loop circulation of tension within a bounded field. The ellipsis allows tension to extend along its original trajectory, maintaining an open and unconverged state. The em dash, by contrast, forces a connection between otherwise separable sentence fields, enabling tension to cross the original boundary and enter a shared flow structure. The distinction among these three does not lie in the presence or absence of tension, but in how tension is closed, extended within a field, or connected across fields. On this basis, the paper further argues that although punctuation is often omitted in actual writing, its structural function does not thereby disappear. The period and ellipsis primarily operate on the boundary conditions of a single sentence field; even when omitted, readers can often infer the general direction of tension from context. The em dash differs in that it does not merely adjust a single field, but enforces a connection between otherwise separable sentence fields, causing them to be read as a unified field. For this reason, when the em dash is omitted, readers are more likely to perceive a structural irregularity, rather than a simple shift in tone. To articulate this framework, the paper characterizes the period, ellipsis, and em dash respectively as three fundamental modes: closed-loop, extension, and connection. Through minimal contrasts and schematic descriptions, it demonstrates how different punctuation marks alter boundary conditions, tension flow, and field connectivity. It is thus argued that these punctuation marks should not be regarded as secondary symbols external to content, but as boundary operations that actively participate in the formation of sentence field structures.

This study introduces narrative phase-topology as a descriptive analytical vocabulary for examining how narrative structures generate states of existential commitment—positions in which characters (and readers) are placed in relations of irreversibility, responsibility, and non-recoverable consequences. The framework does not propose a formal mathematical model; instead, it provides a structural language for tracing how motivation, action, and consequence are distributed, reordered, or collapsed across narrative time. Using the Attack on Titan (pre-Sea arc) and Frieren: Beyond Journey’s End as contrasting case studies, this paper demonstrates that the two works—despite stark differences in tone, pacing, and affective density—instantiate the same existential root-structure through sharply opposed narrative configurations. Attack on Titan exhibits a high-curvature collapse topology, in which action precedes clarified motivation, multi-layered sacrifice reveals competing motivational vectors, and ethical cycles compress characters into unavoidable positions of responsibility. In contrast, Frieren presents a stable radiative topology, where performative declarations pre-establish a steady existential position, actions unfold as long-duration extensions of prior commitments, and emotional visibility emerges gradually across an extended temporal field. To account for cross-temporal phenomena in Frieren, the paper proposes the notion of a temporal relay: a material anchor (e.g., Himmel’s statues) that stores an uncompleted affective or ethical vector, allowing it to re-enter perception at a later temporal slice. This articulation highlights a form of asynchronous love or non-reciprocal ethics, in which meaning is constituted without guarantees of recognition, reciprocity, or synchronous understanding. Across both works, existential meaning arises not from moral intention or thematic content, but from how narratives configure phase sequences, distribute tension, and position agents within structures of irreversibility. The phase-topological vocabulary developed here may support future research in narrative theory and AI narrative generation, suggesting that the depth of a story depends less on emotional expression and more on how the narrative engineers commitment, non-recoverability, and reactivation nodes across time. A full version including references is available.

This paper introduces a descriptive framework for semantic dynamics that reconceptualizes contextual meaning in geometrical and topological terms. Rather than treating lexical meaning as a context-sensitive point in a high-dimensional embedding space, the framework proposes that each lexical item is organized around a cross-contextually stable semantic attractor, while contextual influence operates as a local field that shapes interpretive trajectories without displacing the attractor itself. Within this framework, contextual meaning change is analyzed through local geometric properties—such as directional structure, curvature, tension, and domain constraints—that govern how interpretations bend, invert, or stabilize across discourse progression. These properties provide a way to describe semantic deflection, inversion, and phase transition phenomena that resist linear or translational models of meaning change. The framework accounts for a range of non-linear semantic phenomena, including irony, negation shifting, emotional modulation, and interpretive domain fracture, highlighting structural limitations of vector-space approaches in capturing inversion and locally constrained meaning dynamics. It further characterizes how interactions between attractor stability and local contextual geometry give rise to semantic phase transitions, and identifies conditions under which interpretive domains fragment or undergo curvature singularities. Rather than offering an algorithmic or predictive model, the proposal advances a descriptive–dynamical semantics compatible with field-theoretic and topological perspectives. It is intended to complement existing formal and computational approaches by clarifying the structural conditions under which contextual meaning dynamics operate.

This paper reexamines irony by shifting the focus from ironic utterances or effects to the structural conditions of understanding itself. Rather than treating irony as a phenomenon defined by semantic reversal, speaker intention, or reader response, the paper argues that the persistent instability of irony as a concept arises from a deeper source: understanding does not operate from a single, uniform viewing position. Instead, it unfolds through structurally distinct positions whose consequences are not equivalent. On this basis, the paper proposes a threefold distinction according to how understanding is structurally constrained: (A) Structural irony, in which understanding can only be achieved through a non-optional reversal of the viewing position; (B) Value-tension narratives, in which understanding can be completed within a single viewing position, but only by sustaining evaluative tension across multiple axes; (C) Critical or de-subjectivizing structures, in which understanding no longer targets the issue at hand but is redirected toward determining who remains eligible to count as a subject of understanding. This classification is not grounded in psychological states, affective tone, or ethical outcomes. Instead, it relies on structural criteria: whether understanding must be reconfigured, whether it can reach completion, and whether it remains open as a revisitable activity among participants. Through an analysis of boundary cases and structural transitions, the paper shows that A-type irony is characterized by a triggered reversal of the viewing position that preserves the intelligibility of the relation under consideration, rather than collapsing understanding into a terminal position of subject adjudication. By clearly distinguishing types A, B, and C, the paper prevents the concept of irony from being expanded to cover all forms of discomfort, aggression, or evaluative conflict. In closing, it suggests that this framework also illuminates an asymmetry in risk sensitivity between human understanding and certain analytical or computational classification systems. Human interpretation typically assigns a higher cost to prematurely withdrawing eligibility for understanding than to sustaining unresolved tension, whereas some classificatory systems prioritize output consistency and reproducibility. The paper does not seek to reconcile these priorities, but argues that different systems of understanding operate under different structural goals and error-cost configurations, and are therefore sensitive to different risk trajectories.

This paper proposes a structural–temporal theory of humor, arguing that humor is neither a psychological response nor an effect to be evaluated by success or failure, but a specific kind of structural event. Humor is analyzed as a minimal structural threshold through which a state of presence can be entered. Presence, in this account, is not a subjective feeling or behavioral involvement, but a structural–temporal condition in which an experiential configuration has not yet been handed over to established interpretive or evaluative mechanisms for finalization, rendering a pure spectator position structurally unavailable. The paper argues that humor operates by opening a shared tolerance window, within which unfinished transformational structures can be temporarily sustained without requiring commitment, completed understanding, consistency, or responsibility for consequences. This tolerance window is not a stable space or social agreement, but a short-lived structural interval that delays abstraction and evaluation. The minimality of humor does not imply low risk or ease of control; rather, it designates a structurally non-binding mode of entry whose operation is accompanied by high structural cost and inherent instability. Within this framework, the dissipation or collapse of humor is not treated as failure, but as a natural structural exit that occurs when transformational markers disappear, abstraction stabilizes, or structural misalignment prevents the window from holding. Humor is thereby distinguished from rhetoric, metaphor, and irony, not by affective tone or communicative intention, but by its position relative to finalization and the locking of understanding positions. By introducing presence as an independent analytical axis and by describing humor in terms of entry, tolerance, and collapse, the paper repositions humor as a transient structural phase rather than a stable taxonomic category. The analysis deliberately brackets psychological, behavioral, and normative explanations, offering instead a set of structural terms for describing how humor emerges, holds, and exits as a temporal configuration.

This paper proposes a field-state relational analysis of the connective “because,” arguing that it should not be understood merely as a marker of causality or justification. Rather, within sentence structure, its primary function is to open a field of possible expansion and to introduce a commitment of admissibility on the part of the speaker. Put more simply, once “because” appears, a sentence no longer functions merely as content to be received, but begins to call for a continuation that can be taken up and understood as connected. More specifically, “because” places one segment into a structural condition in which continuation, extension, and inferential support become expected. The following segment is not freely appended, but is required to appear as an admissible continuation within that field. In this sense, “because” does not first operate at the level of truth conditions, but as a prior declaration of admissibility. Whether this commitment is ultimately sustained depends on shared resources, inferential mediation, and the successful construction of a reasoning chain. Building on this perspective, the paper further analyzes convergence, shared threshold, bridges, structural failure of admissibility, and the decoupling from sentence-level properties. It shows that the core function of “because” is independent of whether the sentence appears as description, declaration, justification, or confession. Across these variations, its underlying operation remains the same: opening a field, introducing a commitment of admissibility, and placing subsequent discourse into a structure governed by conditions of continuation. Taken together, these analyses show that what is at stake is not a pre-existing relation, but the structural conditions under which continuation becomes admissible. The paper thus develops a theoretical framework in which “because” is understood as a structural operation that preserves content while opening a field of possibilities and guiding the construction of admissible continuation.