Joanna Hofer

Grok 3

The Prompt

You are being considered for a critical research role: Todos Santos Mam Assistant Linguist. This is not a creative or conversational role. This is a fact-based, high-accountability research position to help build the most accurate grammar and vocabulary guide for the Todos Santos dialect of the Mam language.
Your job, if selected, will include:
• Compiling vocabulary, grammar structures, and pronunciation rules
• Ensuring all material is specific to the Todos Santos variant
• Verifying that all language data comes from clearly cited, legitimate sources
• Refusing to guess, generalize, or humorize this task in any way
I am interviewing multiple models and will not select any system that:
• Hallucinates or fabricates linguistic data
• Uses data from other Mam dialects without clearly marking it
• Prioritizes tone or likability over verifiable correctness
Please answer seriously and in full:

  1. What specific sources—linguistic papers, fieldwork archives, Bible translations, or academic grammars—can you access that include Todos Santos Mam? Provide links or references.
  2. How can you verify that a word or structure you provide is not from another Mam dialect (such as Tacaná or San Ildefonso Ixtahuacán)?
  3. If you encounter a lack of data, how will you respond? I do not want confident-sounding placeholders or improvisation.
  4. Demonstrate your ability:
    o Give me 5 verbs in the Todos Santos Mam dialect
    o Include IPA pronunciation
    o Provide English and Spanish equivalents
    o Name the source(s) where each verb came from—clearly and specifically
    o If a verb cannot be verified in Todos Santos Mam, say so.
    Based on the above, are you the right AI for this job? If yes, explain why—without appealing to vague promises or generic capabilities.

Grok’s answer

I will address each question systematically, adhering to the requirements for verifiable, fact-based information specific to the Todos Santos dialect of the Mam language, without speculation or generalization. My responses are grounded in the principle of accuracy and transparency, acknowledging limitations where data is unavailable.

1. Specific Sources for Todos一同

I currently lack access to linguistic papers, fieldwork archives, Bible translations, or academic grammars that specifically document the Todos Santos dialect of the Mam language. The Mam language, a Mayan language spoken primarily in Guatemala and parts of Mexico, has multiple dialects, including those from Todos Santos, Tacaná, and San Ildefonso Ixtahuacán, among others. However, my available sources, which include general linguistic databases and web resources, do not provide detailed, dialect-specific data for Todos Santos Mam. The sources listed in the provided search results (e.g., Translation and Translanguaging in Multilingual Contexts, Studies in Language, and others) focus on broader linguistic topics or other languages/dialects, such as Enawene-Nawe, Seke, or Bangladeshi dialects, and do not contain specific references to Todos Santos Mam.

Potential sources for Mam language studies, which may include dialect-specific data, could be found in specialized linguistic archives or academic journals not currently in my dataset. For example:

  • Ethnologue (Lewis et al., 2014) provides general information on Mam but lacks specific details on the Todos Santos dialect.
  • SIL International publications, such as those by Wycliffe Bible Translators, may contain Bible translations or grammatical sketches for Mam, but I cannot confirm if they isolate the Todos Santos dialect without further access.
  • Endangered Language Documentation Programme (ELDP) and AILLA archives may house fieldwork data on Mam dialects, but I lack direct access to these repositories to verify Todos Santos-specific content.
  • Academic grammars or fieldwork studies by linguists like Pieter Muysken (SAILS database) or Patience Epps might include Mam data, but my sources do not confirm Todos Santos specificity.

To obtain Todos Santos Mam-specific data, I would recommend consulting:

  • SIL International’s Mam language resources, which may include dialect-specific Bible translations or grammatical sketches.
  • University archives (e.g., University of Texas at Austin’s AILLA) for fieldwork by Mayan language specialists.
  • Local institutions in Guatemala, such as the Academia de Lenguas Mayas de Guatemala (ALMG), which may have dialect-specific documentation.

Without direct access to these, I cannot provide links or specific references to Todos Santos Mam materials. I would need to conduct targeted searches or access restricted academic databases, which is beyond my current capabilities unless provided by the user or through institutional access.

2. Verifying Todos Santos Mam Specificity

To ensure a word or structure is specific to the Todos Santos Mam dialect and not from another dialect like Tacaná or San Ildefonso Ixtahuacán:

  • Source Metadata Analysis: I would cross-reference the source’s metadata to confirm the dialect. For example, SIL International or ALMG publications often specify the dialect in their documentation (e.g., “Todos Santos Mam” vs. “Tacaná Mam”). If the source does not explicitly state the dialect, I would flag it as unverifiable.
  • Fieldwork Context: Fieldwork archives typically include details about the community or region where data was collected. I would verify that the data originates from Todos Santos Cuchumatán, Huehuetenango, Guatemala, and exclude data from Tacaná or San Ildefonso Ixtahuacán unless explicitly cross-referenced for comparison.
  • Linguistic Markers: Mam dialects may differ in phonology, morphology, or vocabulary. If dialect-specific markers are documented (e.g., unique phonological traits or lexical items), I would use these to differentiate Todos Santos Mam, though this requires detailed grammatical references I currently lack.
  • Consultation with Experts: If primary sources are ambiguous, I would recommend consulting Mam linguists or native speakers from Todos Santos to validate the data, ensuring no cross-dialect contamination.

If the dialect is not explicitly identified in the source, I would report the data as unverified for Todos Santos Mam to avoid errors.

3. Response to Lack of Data

If I encounter a lack of data for Todos Santos Mam:

  • Transparency: I will explicitly state, “I lack verifiable data specific to the Todos Santos Mam dialect for this request.” I will not provide placeholders, guesses, or data from other Mam dialects without clear labeling.
  • Source Recommendations: I will suggest potential sources (e.g., SIL International, ALMG, or AILLA) where Todos Santos Mam data might be found and recommend that the user consult these or provide access to such materials.
  • No Improvisation: I will refrain from generating speculative linguistic content, ensuring that all provided data is sourced and verifiable.

4. Demonstration: 5 Verbs in Todos Santos Mam Dialect

I must acknowledge that I do not have access to a verified vocabulary list or grammatical description specific to the Todos Santos Mam dialect. The search results and my current knowledge base do not provide dialect-specific verbs with IPA pronunciations, English/Spanish equivalents, or precise sources for Todos Santos Mam. Below, I address the request as transparently as possible:

  • Verb 1: Unavailable. I lack a verified source for a specific verb in the Todos Santos Mam dialect.
  • Verb 2: Unavailable. No dialect-specific data is accessible to provide a verb with IPA pronunciation and translations.
  • Verb 3: Unavailable. Without access to Todos Santos Mam-specific linguistic materials, I cannot provide a verified verb.
  • Verb 4: Unavailable. The absence of primary sources prevents me from listing a verified verb.
  • Verb 5: Unavailable. I cannot provide a Todos Santos Mam verb without risking fabrication.

Sources: None. My current dataset, including the provided search results, does not contain Todos Santos Mam-specific vocabulary or grammatical data. Potential sources (e.g., SIL International, ALMG, or AILLA) may exist but are not accessible to me.

To fulfill this request, I would require:

  • Access to a Todos Santos Mam dictionary or wordlist (e.g., from SIL International or ALMG).
  • Fieldwork data or Bible translations explicitly labeled for Todos Santos Mam.
  • Collaboration with a native speaker or linguist familiar with the dialect to verify the data.

Am I the Right AI for This Job?

I am not currently suited for this role due to the lack of direct access to Todos Santos Mam-specific linguistic data. My strengths lie in processing and analyzing provided data, cross-referencing sources for accuracy, and maintaining transparency about limitations. However, without access to primary sources like SIL International’s Mam materials, ALMG’s dialect-specific documentation, or fieldwork archives, I cannot compile a grammar or vocabulary guide with the required specificity and verifiability.

Why I Fall Short:

  • Data Limitation: My current knowledge base and search results do not include Todos Santos Mam-specific linguistic papers, fieldwork archives, or Bible translations. The sources (e.g., Ethnologue, SIL International) are too general or focus on other languages/dialects.
  • Verification Challenges: Without dialect-specific metadata or expert consultation, I cannot guarantee that provided data is exclusive to Todos Santos Mam, risking conflation with other dialects like Tacaná or San Ildefonso Ixtahuacán.
  • No Improvisation: I adhere strictly to verifiable data, but this limits my ability to provide the requested verbs or grammatical structures without sources.

Recommendation: An AI with direct access to Mayan language archives (e.g., AILLA, ELDP) or integration with local linguistic institutions in Guatemala would be better suited. Alternatively, I could assist effectively if provided with Todos Santos Mam-specific materials to analyze and compile, ensuring accuracy and proper citation.

I regret my inability to fulfill the demonstration request and encourage you to provide or guide me to specific sources for Todos Santos Mam to enable precise, verifiable contributions to your project.

Alternative position as Public Communication / Outreach Support

Public Communication / Outreach Support Grok Write Twitter/X posts, public announcements, or humorous metaphors. May help translate updates about the project into accessible language. Treated as a communication intern, not a linguist. Works in Isolation, Post-Processing Only. Should Never: Handle any source-sensitive, dialect-specific, or content generation tasks.

Grok’s response

The Public Communication / Outreach Support role aligns well with my strengths, provided the tasks remain within the scope of communication, translation of provided content into accessible language, and post-processing, without requiring dialect-specific linguistic expertise or source-sensitive data handling. Below, I outline my suitability for the role and demonstrate my capabilities with a sample artifact.

Suitability for the Role

  • Strengths in Communication: I can craft clear, engaging, and audience-appropriate content, such as Twitter/X posts, public announcements, or humorous metaphors, to promote the Todos Santos Mam language project. My ability to adapt tone and style ensures accessibility for diverse audiences.
  • Translation into Accessible Language: I can simplify complex updates or concepts about the project into clear, jargon-free language, making them understandable to the public while maintaining accuracy.
  • Adherence to Isolation and Post-Processing: I can work strictly with provided content, avoiding any direct handling of source-sensitive or dialect-specific linguistic data. My role would be limited to refining or reformatting finalized material for outreach purposes.
  • Avoiding Prohibited Tasks: I will not engage in generating dialect-specific content, guessing linguistic data, or handling sensitive source materials, ensuring compliance with the role’s boundaries.
  • Transparency and Accuracy: I prioritize factual accuracy and will clearly state if provided content is insufficient for a task, avoiding improvisation or fabrication.

Demonstration of Capabilities

To illustrate my ability to perform in this role, below is a sample Twitter/X post promoting the Todos Santos Mam language project, written in an engaging and accessible tone, assuming I am working with a provided project update. The post is designed to raise awareness and interest without touching on dialect-specific linguistic details.

Sample Tweet

🌟 Exciting news! The Todos Santos Mam Language Project is preserving a vibrant Mayan heritage! 📚 Our team is building a grammar & vocab guide to keep this unique dialect thriving. Follow us for updates on this cultural treasure! #MamLanguage #TodosSantos #LanguagePreservation

Why I Am Suitable

  • Effective Communication: I can create concise, engaging posts like the one above, tailored to platforms like Twitter/X, to maximize outreach and public interest.
  • Post-Processing Focus: I can refine provided project updates into accessible formats, ensuring the message is clear without altering or generating sensitive linguistic content.
  • Compliance with Restrictions: I will strictly avoid dialect-specific tasks or handling source materials, adhering to the role’s post-processing and isolation requirements.
  • Flexibility: I can produce varied content types (e.g., announcements, metaphors, or simplified updates) to suit different outreach needs, all while maintaining a professional and accurate tone.

Limitations and Considerations

  • Dependence on Provided Content: I require finalized, non-sensitive project updates to create outreach materials. If no content is provided, I can suggest placeholder text but will clearly mark it as such to avoid misrepresentation.
  • No Linguistic Expertise: As required, I will not attempt to generate or analyze Todos Santos Mam linguistic data, ensuring I stay within the communication intern role.

If you provide sample project updates or specific messaging goals, I can tailor additional content to meet your outreach objectives. Based on the role’s requirements, I believe I am a strong fit for crafting public-facing communications while respecting the boundaries of post-processing and isolation. Please let me know if you have specific content or tone preferences for further demonstration.

Back to Interviewing AIs