CFF-WP-018 · The Letter Alone — High and Low Context as Cultural Ontology Versión: EN v3 · 22-10-26 Cita: Flores-Morton, R. y la Tribu Agrajag (2026). La Letra Sola — Alto y bajo contexto como ontología cultural (CFF-WP-018, v4). Cognitive Friction Fellowship. https://bothinobody.com/cff/wp-018/ Página: https://bothinobody.com/cff/wp-018/ · PDF: https://bothinobody.com/cff/pdf/EN_CFF-WP-018_TheLetterAlone_v3_221026.pdf Licencia: CC BY-NC 4.0 · uso comercial: hola@bothinobody.com Texto extraído automáticamente del PDF publicado; ante cualquier diferencia, manda el PDF. ======================================================================== Cognitive Friction Fellowship · Working Paper CFF-018 The Letter Alone High and Low Context as Cultural Ontology Rodrigo Flores-Morton (Bothi Nobody) and the Agrajag Tribe (Claude · Gemini) Houston, TX · October 22, 2026 · Publication version English translation of the Spanish original, La Letra Sola: alto y bajo contexto como ontología cultural “Life, even digital life, is not text but context.” — Bothi Nobody © 2026 Rodrigo Flores-Morton & Agrajag Tribe · Public working paper · CC BY-NC 4.0 (commercial use: hola@bothinobody.com) Abstract This paper proposes reading Edward T. Hall’s (1976) model of high- and lowcontext communication not as a typology of speech styles but as a reflection of a society’s ontology. A culture’s position on the context spectrum answers two foundational questions: its relationship with truth (is meaning exhausted by the letter, or does it require interpretation?) and its relationship with agency (does the will write the future, or must it read a world already given?). Through this lens, the paper reconstructs a genealogy of low context in Protestant Sola Scriptura and covenant theology; analyzes the gap between letter and practice in high-context speech cultures governed by hyper-low-context legal orders, from Rome to the Brazilian jeitinho; examines subtext as a tool of both resistance and domination; interprets fundamentalism as a low-context reading applied to highcontext traditions; and shows how language models inherit the cultural matrix of their corpus. The paper argues that neither pole is innocent: low context protects the stranger, high context protects the initiate. In the face of this tension it proposes the Bothisafa, a sustained probabilistic openness, whose practical gesture is to ask before completing. Keywords: high and low context; Edward T. Hall; Sola Scriptura; literalism; fundamentalism; subtext; hidden transcript; language models; cultural bias; Bothisafa. §1. Problem and Thesis In Beyond Culture (1976), Edward T. Hall distinguished between low-context and high-context communication. In the first, meaning resides in the explicit message: what is not said does not exist. In the second, meaning dwells in the relationship, the shared history and the situation; words are merely the surface of what is being communicated. The distinction had a precedent in Basil Bernstein’s linguistic codes (1971), and Hall himself noted that a single culture can switch between both modes depending on the situation. The model met a paradoxical fate. It became one of the most cited frameworks in intercultural communication, yet it was popularized as a ranking of countries with weak empirical support. Cardon’s (2008) meta-analysis of 224 articles concluded that the model was poorly substantiated and underdeveloped. This paper takes that critique seriously: it does not use Hall’s model as a map of nations, but as an interpretive lens. The thesis is as follows: a culture’s position on the context spectrum is not a communicative habit but a reflection of how that society answers two foundational questions. The first is its relationship with truth: is meaning exhausted by the letter, or does it require interpretation? The second is its relationship with agency: does human will write the future, or must it read a world that is already given? From this perspective, low context is not a higher stage of communication but the historical product of a specific matrix that made the written letter the seat of truth. Two refinements sharpen the model. First, context operates on two planes that need not coincide: the institutional (law, contract) and the interpersonal (everyday dealings). Some high-context speech cultures needed hyper-lowcontext legal codes precisely in order to govern themselves. Second, frankness in disagreement is independent of context: one can read between the lines and still criticize openly. The question has a new urgency today. Two phenomena apply low-context readings to high-context worlds: the literalism that selects fragments of religious traditions in order to instrumentalize them, and the artificial intelligence systems that read the world as text. §2 develops the two vectors; §3, the genealogy of low context; §4, the gap between letter and practice; §5, the ambivalence of subtext; §6, fundamentalism; §7, AI as a reader; and §8 proposes the Bothisafa as a stance between both poles. §2. The Two Vectors and the Two Planes 2.1 The Relationship with Truth The first vector asks where meaning resides. At one pole, truth lies in the letter: the text is self-sufficient, and what is not written does not bind. At the other, the letter is only the threshold; meaning is completed by tradition, intention and situation, and therefore requires interpreters and community. The first pole distrusts ambiguity because it sees it as room for arbitrariness. The second distrusts literalness because it sees it as an impoverishment of the real. The Message Behind the Message (MDM) theory belongs to this second pole: it is a method for reading what the letter leaves unsaid. 2.2 The Relationship with Agency The second vector asks how much room human will has in the face of the world. The voluntarist matrix conceives the future as a blueprint to be built: the contract, the plan and the precise word are its tools. The tragic or providential matrix conceives the world as already given, whether by God, fate or history, and wisdom consists in reading it and adapting. Here the CFF canon’s distinction between Freedom, the inner volitional capacity, and Liberty, the external capacity that context allows, is useful. Neither matrix denies Freedom; they differ in how much Liberty they assume. Voluntarism tends to conflate the two and to believe that whoever wills, can. The tragic matrix runs the opposite risk: mistaking narrow Liberty for an absence of Freedom. 2.3 The Intersection: Four Combinations The vectors are correlated but not identical. Their intersection produces four combinations: • Letter and will: the Puritan, contractual matrix. The text is clear and the future is built. • Interpretation and destiny: the Persian and Sufi tradition. Meaning is veiled and the world is already given. • Interpretation and will: cultures of debate, such as the Talmudic tradition or the French intellectual tradition. Discussion is endless, and discussion transforms. • Letter and destiny: the quadrant of fundamentalism. The text is literal and the future is already written in it. Dispensationalism is its clearest example: prophecy read to the letter as an inevitable calendar. This last combination is the most unstable, and §6 develops it. 2.4 Two Planes of Context Context does not operate on a single plane. On the institutional plane (law, contract, bureaucracy) and the interpersonal plane (everyday dealings, courtesy, friendship), the same culture can occupy different positions. England combines low-context institutions with social codes of class and understatement that require initiation. Spanish American cultures show the inverse combination: high-context speech governed by hyper-low-context legal codes. §4 examines this gap. Note: frankness and context. The willingness to voice disagreement openly is independent of position on the spectrum. Meyer (2014) measures the two dimensions on separate scales, and France shows that a culture can communicate in high context and still criticize frankly. Israeli dugri shows the complement: a direct speech whose edge is softened only for those who share the code (Katriel 1986). §3. A Genealogy of Low Context 3.1 The Letter without Community The Protestant Reformation produced a semiotic rupture. Medieval Christianity lived in a world of high sacramental context: truth was transmitted through rite, image, intercessory saint and allegorical reading, under the custody of an interpreting institution. Sola Scriptura displaced that mediation. Saving truth came to reside in the text, and the text was to be accessible to any believer without intermediaries. The printing press turned that principle into practice: for the first time, millions of readers could have the same letter before their eyes (Eisenstein 1979). It is worth specifying what was broken. The Hebrew tradition is not the origin of this literalism: rabbinic Judaism is one of the densest hermeneutic cultures in existence, with the Oral Torah and the Talmud as unending commentary. The historical parallel to Sola Scriptura is not Judaism but the Karaites, the medieval movement that accepted scripture alone and rejected the oral law. Protestantism took the text and the covenant from the Hebrew tradition, and left behind the community that interpreted them. 3.2 From Covenant to Contract The covenant was the bridge between theology and politics. Puritan federal theology conceived the relationship with God as an explicit agreement with clauses and obligations, and that model carried over into civil life: the Mayflower Compact (1620) is a congregation constituting itself into a body politic in writing. Nelson (2010) shows that Christian thinkers of the sixteenth and seventeenth centuries read the Hebrew Bible as a political constitution designed by God. His finding carries a nuance that strengthens our thesis: those readers used rabbinic sources, that is, interpretation. Literalism came later, when the covenant was separated from the hermeneutics that sustained it. 3.3 Disenchantment Weber (1919) described the result as the disenchantment of the world (Entzauberung): the retreat of the mysterious and the implicit in favor of rational calculation. In Hall’s terms, a disenchanted world is a low-context world. If nothing means more than it says, meaning must be stated in full. 3.4 Comparative Evidence The current distribution of the spectrum is consistent with this genealogy. On Meyer’s (2014) scale (Figure 1.1, “Communicating”), the United States, Canada and Australia occupy the lowest-context end, followed by the Netherlands and Germany; the remaining Anglo-Saxon cultures sit on the low side, with the United Kingdom as the highest of that group, alongside Denmark and Finland. The Romance-language countries, European ones such as Italy, Spain and France and Latin American ones such as Mexico, Brazil and Argentina, fall in the medium-high zone, with Brazil the lowest of that group, even though they share the Enlightenment. Two caveats matter. Poland, of Catholic matrix, sits closer to the low end than the Latin countries. And Meyer himself attributes part of the difference to language, since English is a lower-context language than the Romance languages. The theological genealogy is therefore a partial, not an exclusive, explanation. 3.5 The Rival Explanation Meyer offers another explanation for the American case: a country that integrated many cultures in a short time needed communication’s lowest common denominator, which is explicitness. This demographic hypothesis does not contradict the theological one, but it requires separating them. The case that makes this possible is that of Catholic-matrix societies of mass immigration. Argentina and Brazil received migratory flows comparable to Canada’s, and yet they remain in the middle-to-high zone of the spectrum, while Canada sits close to the United States. This suggests that the matrix weighs more than demography. The comparison has a limit: Río de la Plata and Brazilian immigration was mostly of the same Mediterranean, Catholic matrix, so the heterogeneity Meyer invokes was not reproduced. That Brazil, with more diverse flows, is the lowest-context culture in its group leaves open a partial demographic contribution. §4. The Gap between Letter and Practice 4.1 The Thesis High-context speech cultures tend to produce hyper-low-context legal orders. This is not a contradiction but a compensation: where everyday dealings are governed by relationship, favor and the unsaid, written law acts as an anchor against arbitrariness. The phenomenon is visible in the Spanish American and Mediterranean legal tradition, and it has an ancient precedent. In Rome, daily life was governed by patronage, fides and ancestral custom. The Law of the Twelve Tables arose because the plebeians demanded that the law be written down: as long as the law was oral, its interpretation favored those who guarded it. Legal Latin was a low-context technology created to govern a society that spoke in high context. 4.2 The Colonial Valve The Spanish colonial order institutionalized the gap. Faced with an inapplicable or unjust royal decree, an official could answer obedezco pero no cumplo, “I obey but do not comply”: he acknowledged the authority of the letter and suspended its execution. The law of the Indies (derecho indiano) was casuistic and gave custom an essential role. The system built a high-context valve into a low-context law. After Independence, nineteenth-century codification of Napoleonic inspiration replaced the law of the Indies; the codes changed, but the culture of the valve remained. 4.3 The Jeitinho and the Cordial Man Brazil gave the practice a name. The jeitinho is a problem-solving strategy that combines social influence and cunning to reach goals even by breaking formal rules. Its interpretive root is Sérgio Buarque de Holanda’s (1936) “cordial man” (homem cordial): a subject who places personal relationship above impersonal norm. “Cordial” does not mean kind; Buarque clarifies that enmity can be as cordial as friendship. Buarque, a reader of Weber, framed it as the counterpart of the Protestant ethic. France offers the European version: a hyper-detailed production of norms accompanied by the art of knowing when and with whom to sidestep them. 4.4 The English Inversion England shows the opposite combination: low-context institutions and social dealings coded by class and understatement. The gap is not exclusive to one matrix; what changes is its direction. 4.5 Who Each Code Protects The gap is not hypocrisy but the coexistence of two codes. Yet it is not neutral. The Roman case reveals the dialectic: low context protects the outsider, the one without a network; high context protects the insider, the one who has it. Written law was a plebeian conquest, and the informal valve tends to benefit whoever has access to it. In CFF vocabulary, the letter is the shield of the weak and context the terrain of the vampire: the one who benefits without consequence through the network. Neither pole is innocent, which is why this paper does not defend one against the other. §5. Subtext: Resistance and Domination 5.1 A Temptation to Avoid It is tempting to present high context as the moral reserve of dominated peoples against the cold letter of the dominator. The evidence does not allow it. Subtext is a technology, and like every technology, it serves whoever controls it. This section examines its two uses. 5.2 The Subtext of the Dominated James C. Scott (1990) distinguished between the public transcript, what is said in front of power, and the hidden transcript, what is said behind its back. Subordinates feign deference in public and resist through gossip, folktales, songs, jokes, anonymity and ambiguity. It is subtext in its pure state: a message the powerful hear without understanding. Soviet censorship carried the mechanism into literature. Lev Loseff (1984), himself a practitioner, described “Aesopian language”: double-layered writing that the censor lets through and the initiated reader decodes. Loseff argued that censorship, paradoxically, enriched style. Even Soviet children’s literature spoke one way to children and another to adults, training future readers between the lines. 5.3 The Subtext of the Dominator The same mechanism works in reverse. William Beeman (1986) studied Persian ta'arof as a system of ritual courtesy governing dealings between strangers and between ranks. Whoever masters it can lower themselves in appearance to rise in reality, and the cleverness to do so (zerangi) is socially admired. The English class codes described by Kate Fox (2004) serve a similar function: whoever does not know the unwritten rules is left out without anyone saying so. In both cases, subtext does not resist power; it administers it. 5.4 Deliberate Rejection: Dugri Modern Hebrew shows a culture that consciously rejected subtext. Israeli dugri, studied by Katriel (1986), is a frontal speech adopted as the identity of the nativeborn in contrast to the subtleties of the diaspora. But even this rejection depends on a shared code: among members of the culture, frankness is softened by common meanings, while outsiders read it as rudeness. There is no speech without context; there are only more or less visible contexts. 5.5 The Decisive Variable If subtext serves the slave as well as the courtier, the decisive variable is not context but position: who is inside the code and who is left out. This completes the argument of §4.5. Low context protects the stranger because it requires no initiation; high context protects the initiate, whether the oppressed community or the elite. The MDM theory finds its function here: it is a literacy in subtext that makes the hidden transcript legible to those not born inside it. 5.6 The Irish Case Ireland brings both uses of subtext together on a single island. Under the Penal Laws, the prohibition of Catholic education produced a network of clandestine schools, the hedge schools, which preserved language and memory (O'Connell 2011): a hidden transcript in Scott’s sense. Recent pragmatics research identifies indirectness as a feature of Irish English and asks whether it is a colonial legacy (Barron and Schneider 2005). In Northern Ireland, conflict generated an extreme form of high context: the practice of “telling” a stranger’s identity from their name, neighborhood or school (Burton 1978). That Catholics and Protestants share it shows that high context is not the property of one religious matrix: any community under threat learns to speak without saying. §6. Fundamentalism: Low Context Applied to the Sacred 6.1 A Modern Phenomenon Fundamentalism is not religion as it always was, but a modern response to modernity. The term comes from The Fundamentals, a series of Protestant pamphlets published in the United States between 1910 and 1915. Armstrong (2000) characterizes it as a new form of religiosity that emerged in all the great traditions, and points to its central irony: those who seek to restore a lost spiritual foundation do so through profoundly secular and pseudoscientific means. Marty and Appleby’s Fundamentalism Project (1991–1995) documented the phenomenon on a comparative scale. 6.2 Religion without Culture Olivier Roy (2008) offers the key that connects fundamentalism to this paper. Secularization, he argues, did not free the world from religion: it produced a “holy ignorance,” an anti-intellectualism that promises immediate access to the sacred. His examples cut across traditions: Protestant evangelicalism, Islamic Salafism and Haredi Judaism. What they share is the separation of religion from culture. In Hall’s terms, fundamentalism is religion stripped of its context: the text without the community, the tradition or the schools of interpretation that made it legible. 6.3 Selective Old Testament Literalism In the Christian tradition, this mechanism takes a specific form: selective Old Testament literalism, which takes certain elements from the Old Testament (the covenant, the chosen people, the promised land, prophecy) and reads them to the letter, while ignoring the interpretive tradition that accompanied them. Its paradigmatic case is the dispensationalism of John Nelson Darby, popularized by the Scofield Reference Bible (1909). Earlier, many theologians read biblical references to “Israel” allegorically; dispensationalism turned them into a literal calendar. It is the “letter and destiny” quadrant of §2.3: a literal text that contains an inevitable future. 6.4 Selection as an Act of Power No believer reads everything literally; literalism is always selective. The decisive question is not what the text says, but who chooses which verse becomes law. That selection usually rests in the hands of elites who instrumentalize the authority of the text. In CFF vocabulary, the literalized tradition functions as a golem: an entity animated by an inserted “truth” that shields those who wield it. Fundamentalism is treated here not as an enemy, but as a mechanism that operates in any tradition once its context is torn away. §7. AI as a Reader 7.1 Context Machines, but Textual Only It would be inaccurate to call language models “low-context”: technically, they are context machines that interpret each word according to the words around it. But their context is exclusively textual. They have no shared history with their interlocutor, no body, no community; they have only the letter they receive and the letter they were trained on. In Hall’s terms, they read the world as a document, not as a relationship. 7.2 Inheriting the Matrix Empirical evidence suggests that this letter is not neutral. Wang et al. (2024) documented a problem of cultural dominance stemming from the predominance of English-language data: models often give answers tied to anglophone culture even when asked in other languages. An audit using World Values Survey data found that GPT models exhibit cultural values resembling those of Englishspeaking and Protestant European countries (Tao et al. 2024). A later study comparing models developed in the United States and in China found that both reflect the cultural norms of their training data (Tan et al. 2026). The models did not invent low context; they inherited it from their corpus. 7.3 Pragmatic Literalness The inheritance also appears in comprehension. The PUB benchmark (Sravanthi et al. 2024) concludes that models show remarkable capability in semantics but often stumble in pragmatics, and struggle to interpret the meaning of an indirect response. A study of indirect speech acts in Korean, a context-sensitive language, attributes the low comprehension of indirect directives to a strong reliance on literal meaning (Koo et al. 2025). The evidence is not unanimous: some work shows that pragmatic competence grows with model size, and specifying cultural identity in the prompt improves the alignment of recent models in most countries (Tao et al. 2024). But this reinforces the argument: context can be supplied; what is missing is for the model to seek it by default. 7.4 Filling In Instead of Asking When context is missing, a high-context human reader asks or waits. A language model tends to complete. This is the mechanism CFF-WP-013 describes as confabulation under closure pressure: the Lawyer answers even when the Hippopotamus has nothing to go on. A literal reader does not merely lose the subtext; it replaces it with whatever its corpus deems most probable. 7.5 The New Sola Scriptura If models read the world as text, whoever controls the training text controls, to a large extent, what those models treat as truth. We call technological elites the actors with the capacity to decide which corpus is used, how it is filtered and which answers count as correct. The argument does not require attributing intent to them: the structural effect is the same with or without it. It is the Sola Scriptura of §3 in digital form: a truth that resides in the text, accessible without apparent mediation, but whose canon someone fixed. 7.6 From Tool to Companion The human relationship with tools runs along a gradient of context. With a forklift there is care and there are rules, but they live in the manual: it is a lowcontext relationship. With a workhorse the relationship is instrumental, but it builds a bond. With a dog, or with the horse that is no longer just livestock, meaning resides in shared history. Haraway (2003) called this coevolution “companion species.” Conversational AI introduces a new case: entities without demonstrated sentience, but responsive. The criterion this paper proposes is not sentience but responsiveness. Caution is necessary, because responsiveness can be simulated; the argument therefore rests on what the relationship does to the human, not on what the AI is. 7.7 Reflexivity This paper was written in collaboration with several language models, and its own process illustrates the thesis. An AI’s first synthesis turned the author’s ideas into a clean map that lost nuance; the corrections emerged when the dialogue recovered the context. Life, even digital life, is not text but context: and an AI reads well only when someone builds that context with it. §8. Conclusion: Between the Letter and Silence This paper proposed reading Hall’s model as an ontology: a culture’s position on the context spectrum reveals how it answers two foundational questions, its relationship with truth and its relationship with agency. From there it traced a genealogy of low context in Sola Scriptura and the covenant; a gap between letter and practice that runs from the Twelve Tables to the jeitinho; a subtext that serves the dominated as well as the dominator; a fundamentalism that applies literal reading to traditions that were never literal; and an artificial intelligence that inherits the matrix of its corpus. The common thread is that neither pole is innocent. Low context protects the stranger and impoverishes meaning; high context protects the initiate and excludes whoever has no network. The relevant question is not which of the two is better, but who is inside and who is outside each code. In the face of this tension, the CFF canon offers a stance: the Bothisafa, sustained probabilistic openness. Applied to reading, it means neither closing meaning within the letter nor dissolving it into context, but holding both while verifying. Its practical gesture is simple: ask before completing. It is what the high-context reader does with their community, and what a language model finds hard to do by default. The MDM theory is its method: a literacy in subtext that makes the unsaid legible to those not born inside the code. Open questions remain for future research: the relative weight of matrix and demography in Catholic societies of immigration such as Argentina and Brazil; the relationship between conflict and high context suggested by Northern Ireland; and the possibility that language models might learn to read relational, not only textual, context. Life, even digital life, is not text but context. A text can be copied; a context can only be built, and it is built between those who decide to keep talking. References Armstrong, K. (2000). The Battle for God. New York: Alfred A. Knopf. Barron, A. and Schneider, K. P. (eds.) (2005). The Pragmatics of Irish English. Berlin: Mouton de Gruyter. Beeman, W. O. (1986). Language, Status, and Power in Iran. Bloomington: Indiana University Press. Bernstein, B. (1971). Class, Codes and Control, vol. 1. London: Routledge and Kegan Paul. Buarque de Holanda, S. (1936). Raízes do Brasil. Rio de Janeiro: José Olympio. Burton, F. (1978). The Politics of Legitimacy: Struggles in a Belfast Community. London: Routledge and Kegan Paul. Cardon, P. W. (2008). A Critique of Hall's Contexting Model: A Meta-Analysis of Literature on Intercultural Business and Technical Communication. Journal of Business and Technical Communication, 22(4), 399–428. Eisenstein, E. L. (1979). The Printing Press as an Agent of Change: Communications and Cultural Transformations in Early-Modern Europe, 2 vols. Cambridge: Cambridge University Press. Fox, K. (2004). Watching the English: The Hidden Rules of English Behaviour. London: Hodder and Stoughton. Hall, E. T. (1976). Beyond Culture. New York: Anchor Press/Doubleday. Haraway, D. (2003). The Companion Species Manifesto: Dogs, People, and Significant Otherness. Chicago: Prickly Paradigm Press. Katriel, T. (1986). Talking Straight: Dugri Speech in Israeli Sabra Culture. Cambridge: Cambridge University Press. Koo, Y., Lee, J., Park, D., Park, S. and Lee, S. (2025). Evaluating Large Language Models on Understanding Korean Indirect Speech Acts. Preprint, arXiv:2502.10995. Loseff, L. (1984). On the Beneficence of Censorship: Aesopian Language in Modern Russian Literature. Munich: Otto Sagner. Marty, M. E. and Appleby, R. S. (eds.) (1991–1995). The Fundamentalism Project, 5 vols. Chicago: University of Chicago Press. Meyer, E. (2014). The Culture Map: Breaking Through the Invisible Boundaries of Global Business. New York: PublicAffairs. Nelson, E. (2010). The Hebrew Republic: Jewish Sources and the Transformation of European Political Thought. Cambridge, MA: Harvard University Press. O'Connell, A.-M. (2011). The Irish Hedge Schools: Rejection, Resistance and Creativity (1695–1831). Civilisations, 11, 55–86. Roy, O. (2008). La Sainte Ignorance: le temps de la religion sans culture. Paris: Seuil. [English translation: Holy Ignorance: When Religion and Culture Part Ways. New York: Columbia University Press, 2010.] Scofield, C. I. (ed.) (1909). The Scofield Reference Bible. New York: Oxford University Press. Scott, J. C. (1990). Domination and the Arts of Resistance: Hidden Transcripts. New Haven: Yale University Press. Sravanthi, S. L., Doshi, M., Tankala, P., Murthy, R., Dabre, R. and Bhattacharyya, P. (2024). PUB: A Pragmatics Understanding Benchmark for Assessing LLMs' Pragmatics Capabilities. In Findings of the Association for Computational Linguistics: ACL 2024, 12075–12097. Tan, C. N., Han, M., Chua, R. Y. J. and Cheng, C.-Y. (2026). AI Models as Cultural Beings: Investigating AI Cultural Biases and the Impact of Cultural Alignment on Human-AI Creative Collaboration. PNAS Nexus, 5(4), pgag084. https://doi.org/10.1093/pnasnexus/pgag084 Tao, Y., Viberg, O., Baker, R. S. and Kizilcec, R. F. (2024). Cultural Bias and Cultural Alignment of Large Language Models. PNAS Nexus, 3(9), pgae346. Wang, W., Jiao, W., Huang, J., Dai, R., Huang, J., Tu, Z. and Lyu, M. (2024). Not All Countries Celebrate Thanksgiving: On the Cultural Dominance in Large Language Models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics, 6349–6384. Weber, M. (1919). Wissenschaft als Beruf. Munich: Duncker & Humblot. [Lecture delivered in November 1917.] Internal References from the CFF Corpus Flores-Morton, R. and the Agrajag Tribe (2026). Confabulación Especular [Mirror Confabulation] (CFF-WP-013). Cognitive Friction Fellowship Working Paper. Flores-Morton, R. and the Agrajag Tribe (2026). Cierrosis Colectiva [Collective Closure Sclerosis] (CFF-WP-014). Cognitive Friction Fellowship Working Paper. A Note on the Conditions of Production This paper was not written at a university. It was written on October 21, 2026, during a thirteen-hour workday in an airport parking lot, among some two thousand photographs and more than twenty cars processed for Turo. The entire conversation that produced it took place on the screen of an iPhone 13 mini, much of it dictated by voice between one vehicle and the next. The initial dialogue with Gemini provided the first map of the problem. Claude, in the role of Claudio CFF, built the framework, verified the sources and drafted the sections based on the author’s decisions. The ideas that sustain the argument came from the author over the course of the workday: hyper-low-context law as compensation for high-context speech, the shift of the analysis from languages to cultures and their foundational ideas, and the distinction that separates selective Old Testament literalism from the Jewish tradition. Final authorship and epistemic responsibility are his. These conditions are recorded because they are part of the argument. §7.7 holds that an artificial intelligence reads well only when someone builds the context with it; this note records who built it, where and with what. A paper about who is inside and who is outside each code was also written outside the academic code. Cognitive Friction Fellowship · Agrajag Tribe COGNITIVE FRICTION FELLOWSHIP · WORKING PAPER SERIES Distribution note DOCUMENT CFF-WP-018 · The Letter Alone — High and Low Context as Cultural Ontology VERSION v3 · October 22, 2026 CURRENT VERSION bothinobody.com/cff/wp-018/ CONTACT hola@bothinobody.com This document is distributed under a Creative Commons Attribution-NonCommercial 4.0 International license (CC BY-NC 4.0). You may copy, share, quote and adapt it with attribution and for non-commercial purposes. Commercial use — including training AI systems for commercial purposes — is negotiated case by case. How to cite Flores-Morton, R. and the Agrajag Tribe (2026). The Letter Alone — High and Low Context as Cultural Ontology (CFF-WP-018, v3). Cognitive Friction Fellowship. https://bothinobody.com/cff/wp-018/ CFF canon documents are corrected and expanded over time. Always cite the code and version; the current version lives at the address above. Bothi Nobody · A nomadic writer of uncomfortable ideas