Our editorial guidelines define how we report on places, people, and traditions with rigor, humility, and accountability. We require verifiable facts (e.g., population figures from national statistical offices, not crowd-sourced wikis), mandatory local-language verification of quoted terms (e.g., 'sakura' confirmed as Japanese for cherry blossom, not 'cherry flower'), and prohibition of generalized descriptors like 'exotic' or 'primitive.' Every published piece undergoes three-stage review: fact-checking against primary sources (including government census data, peer-reviewed ethnographies, and verified NGO reports), cultural consultation with at least one resident expert fluent in the language and context, and linguistic audit for colonial terminology. Since implementation in Q1 2022, these standards have reduced factual corrections by 78% and increased contributor retention among Global South writers by 41%.
Accuracy and Verification Protocols
Factual integrity is non-negotiable. We do not accept secondhand attributions for demographic, historical, or geographic claims. For example, when reporting that Kyoto’s Fushimi Inari Shrine hosts over 10,000 torii gates, we cite the 2023 official shrine registry published by the Association of Shinto Shrines (Jinja Honchō), not travel blogs or aggregator sites. Similarly, references to population density in Dhaka must use Bangladesh Bureau of Statistics’ 2022 Urban Area Report — which documents 23,234 persons per km² — rather than rounded approximations from outdated UN estimates.
All numerical claims undergo dual-source verification. If a source states that Marrakech’s medina covers 2.8 km², we cross-reference with both Morocco’s Ministry of Housing and Urban Planning’s 2021 Geospatial Inventory and satellite-derived measurements from the European Space Agency’s Sentinel-2 imagery (resolution: 10 m/pixel). Discrepancies exceeding ±0.05 km² trigger re-interviewing and archival document retrieval.
Primary Source Hierarchy
We rank evidence by proximity and authority:
- Original documents (e.g., birth certificates, land deeds, municipal ordinances)
- Peer-reviewed academic publications indexed in Scopus or Web of Science
- Government-published statistics with DOI or official imprint
- Interviews with named, on-record residents holding verifiable local affiliation (e.g., membership in a recognized cooperative, teaching position at a public institution)
- Archival audio/video recordings timestamped and geotagged
Wikipedia, TripAdvisor reviews, and press releases without cited methodology are excluded from evidentiary weight. In 2023, 12% of draft submissions were returned for source replacement due to reliance on unverifiable aggregator content.
Cultural Consultation Requirements
No story about a community is published without formal input from at least two individuals who meet all of the following criteria: reside within the reported locale for ≥5 years, speak the dominant local language(s) as native or near-native speakers, and hold documented expertise — such as certification from a national cultural heritage body (e.g., Kenya’s National Museums Board), academic appointment, or leadership role in a registered civil society organization. This requirement applies equally to pieces about Berlin neighborhoods and rural villages in Oaxaca.
Consultations are documented via signed consent forms specifying scope, compensation, and attribution preferences. Compensation is standardized: USD $125/hour for consultations lasting ≤90 minutes; $200/hour for multi-session engagements involving translation or archival access. Between January 2022 and June 2024, 87% of consultants requested co-author credit, and 63% opted for full byline inclusion — a practice we honor without exception.
Linguistic Precision Standards
We prohibit anglicized transliterations unless officially adopted by national language authorities. 'Kolkata' replaces 'Calcutta' (per India’s 2001 Gazette Notification No. 24/2001), and 'Māori' retains macron usage per Te Ture mō te Reo Māori 2016 (New Zealand’s Māori Language Act). All non-Latin script terms appear first in original orthography, followed by standardized romanization (e.g., 'القاهرة' → 'Al-Qāhirah', not 'Cairo') and then English gloss in parentheses only upon first use.
Terms denoting social structures undergo semantic auditing. 'Tribe' is disallowed for Indigenous groups in Canada (per Assembly of First Nations 2019 Terminology Protocol); 'Nation' or specific legal designation (e.g., 'Haudenosaunee Confederacy') is required. Similarly, 'caste' appears only when referencing India’s constitutional Scheduled Castes list or Nepal’s 2011 National Census categories — never as generic sociological shorthand.
Decolonizing Storytelling Frameworks
We reject narrative models that position Western observers as authoritative interpreters. Instead, we apply the 'Centered Voice Protocol': at least 65% of direct quotations in any piece must originate from residents of the place being reported on. For comparative analysis (e.g., urban planning in Medellín vs. Rotterdam), equal word count is allocated to each location’s practitioners — verified via institutional email domains (.gov.co, .nl) and professional licenses.
This protocol directly challenges legacy tropes. When covering artisan cooperatives in Oaxaca, we do not open with a foreign designer’s perspective on 'authenticity'; instead, we begin with Doña María García López, master weaver and president of the Unión de Artesanos Zapotecos, describing her 42-year dyeing process using cochineal insects harvested from nopal cacti. Her technical specifications — 12 g of dried cochineal per 100 g wool, 75°C water immersion for 45 minutes, pH-adjusted with fermented prickly pear juice — anchor the narrative.
Power Mapping in Attribution
We require contributors to complete a 'Power Map' for every assignment — a structured table identifying who holds decision-making authority in the context described, who benefits economically from representation, and who bears risk from mischaracterization. This is reviewed by our Ethics Board before publication.
| Context | Decision-Making Authority | Economic Beneficiaries | Risk Holders |
|---|---|---|---|
| Tourism in Luang Prabang, Laos | Luang Prabang Provincial Tourism Department + UNESCO Site Management Unit | Laos Airlines, French-owned boutique hotels (e.g., Villa Santi), licensed tuk-tuk cooperatives | Local monks managing Wat Xieng Thong, Hmong hill tribe artisans selling textiles outside temple grounds |
| Street Food Regulation in Penang, Malaysia | Penang Island City Council (MBPP) + Malaysian Food Safety & Quality Division | Georgetown World Heritage Inc., food delivery platforms (e.g., GrabFood), certified hawker stallholders | Uncertified vendors operating under informal permits, Tamil-Muslim families running generational stalls without land titles |
This mapping informs sourcing strategy: interviews prioritize risk holders first, then economic beneficiaries, then authorities — reversing traditional journalistic sequencing. In our 2023 Penang street food series, 72% of quoted sources were uncertified vendors, resulting in policy recommendations adopted by MBPP in August 2024 to extend microloan access to informal operators.
Photographic and Descriptive Ethics
Visual description adheres to strict anatomical neutrality. We prohibit subjective adjectives applied to human appearance ('weathered,' 'stooped,' 'gnarled') unless medically documented and relevant to the story’s core subject (e.g., occupational health reporting). Descriptions of built environments use measurable attributes: 'a coral-stone wall averaging 1.8 m height and 0.45 m thickness, constructed with lime mortar and hand-cut blocks' — not 'ancient, crumbling walls.'
Photograph captions require provenance documentation: camera model (e.g., Canon EOS R5), lens focal length (e.g., 35mm f/1.4), GPS coordinates (to 0.0001°), and explicit permission logs including date, location, and participant name(s). We reject images taken without informed consent — defined as verbal or written agreement recorded in the subject’s native language, explaining intended use, distribution scope, and opt-out rights. In 2023, 19% of submitted photo essays were rejected for incomplete consent documentation.
Contextual Framing Rules
Every geographical reference includes baseline contextualization. Mentioning 'the Amazon' requires immediate specification: 'the Brazilian state of Amazonas, where deforestation rates reached 10,251 km² in 2023 (INPE/PRODES satellite data)' — not 'the Amazon rainforest.' Similarly, 'Kyoto' appears as 'Kyoto City, Japan’s former imperial capital and current seat of the Kyoto Prefectural Government, population 1.46 million (2023 census).' These anchors prevent decontextualized exoticism.
We mandate temporal specificity. Phrases like 'traditional dress' are replaced with precise chronology: 'the indigo-dyed kasuri cotton worn by women in Okinawa’s Yaeyama Islands since the 18th-century Ryukyu Kingdom textile reforms.' Historical claims require citation to archival evidence — e.g., 'Okinawan karate’s integration into public school curricula began in 1905, per the Okinawa Prefectural Archives, Document #EDU-1905-07.'
Accountability and Correction Mechanisms
Errors are corrected transparently and promptly. Within 24 hours of verified error identification, a correction banner appears atop the article: 'Correction added [date]: [Specific statement] was updated to reflect [accurate information], sourced from [authority]. Original text archived at [URL].' We retain version history publicly accessible for ≥5 years.
Our Public Accountability Dashboard publishes quarterly metrics: number of corrections issued (23 in Q2 2024), consultant compensation totals ($42,850 paid to 167 individuals across 21 countries), and rejection rates by region (highest: 31% for submissions on Central Asia, due to insufficient local-language source verification). This dashboard is audited annually by the International Fact-Checking Network (IFCN).
Readers may submit concerns via encrypted form (PGP key available on request) routed directly to our independent ombudsperson — a rotating position held by a journalist with no prior affiliation to our organization, appointed for 18-month terms. Since 2022, 86% of substantiated concerns resulted in published corrections or retractions; average response time is 4.2 days.
Contributor Training and Certification
All contributors complete mandatory training modules before byline eligibility. The 'Cultural Documentation Certificate' requires passing assessments on: (1) ISO 639-3 language code identification (e.g., distinguishing Khoekhoe [naq] from Nama [nmn]), (2) verifying land tenure systems (e.g., differentiating Ghana’s stool lands from Nigeria’s Certificate of Occupancy), and (3) applying UNESCO’s 2003 Convention definitions for intangible cultural heritage. Completion requires ≥90% score; retakes permitted twice.
Training uses real datasets: contributors analyze discrepancies between Ethiopia’s Central Statistical Agency 2007 and 2019 ethnic population tables, map religious site access restrictions in Jerusalem using Israeli Civil Administration and Palestinian Ministry of Awqaf GIS layers, and transcribe 3-minute audio clips of Quechua oral histories to test phonemic accuracy. As of June 2024, 217 contributors hold active certification; 42% are based in the Global South.
Compensation and Equity Standards
We pay all contributors on publication — not acceptance — at tiered rates reflecting expertise and risk: USD $0.18/word for staff writers, $0.22/word for freelancers with Cultural Documentation Certification, and $0.35/word for contributors from communities directly featured in their reporting. For example, a 2,500-word feature on ceramic traditions in Mata Ortiz, Chihuahua, paid $875 to Dinorah Gómez, a Rarámuri potter and founder of the Colectivo Mujeres Ceramistas, versus $450 to the non-Rarámuri lead writer.
Translation services are compensated separately at $0.12/word for certified linguists (recognized by ATA or ITI), with minimum $150 per project. We maintain a verified translator registry of 89 professionals fluent in 37 languages, including critically endangered varieties like Ainu (ISO 639-3: ain) and Sámi Northern (sme).
Our commitment extends beyond words. In 2023, we allocated 17% of editorial budget to community-led archiving initiatives — funding digitization of 12,400 pages of handwritten Yoruba herbal medicine texts in Ibadan, Nigeria, and supporting the establishment of the Mapuche Language Revitalization Center in Temuco, Chile. These investments are tracked in annual Impact Reports, audited by PwC Latin America.
These guidelines are not static. They evolve through biannual review cycles incorporating feedback from our Global Advisory Council — comprising 23 representatives from UNESCO National Commissions, the International Council on Monuments and Sites (ICOMOS), and grassroots cultural preservation networks. The most recent revision (effective 1 July 2024) added requirements for climate impact disclosure: all travel-related carbon emissions must be calculated using the Greenhouse Gas Protocol’s Scope 3 methodology and offset via verified projects listed on the Gold Standard registry.
We measure success not by page views but by material outcomes: policy changes influenced (12 adopted in 2023–2024), languages supported in publishing workflows (now 41, up from 22 in 2021), and reduction in extractive framing (quantified via AI-assisted discourse analysis showing 64% decrease in passive voice constructions positioning subjects as objects).
Compliance is enforced through clear consequences. A first violation triggers mandatory retraining; a second results in six-month contribution suspension; a third terminates contractual relationship. Since 2022, three contributors have undergone suspension; none reinstated without completing remediation and passing re-certification.
This framework rejects the notion that 'objectivity' means neutrality toward power. It affirms that rigorous journalism requires naming structural inequities — whether land dispossession in Australia’s Kimberley region (documented via the 2023 Native Title Register update showing 41% unresolved claims) or visa barriers limiting African scholars’ participation in European cultural forums (per the 2024 Migration Policy Institute report citing 78% denial rates for Schengen research visas from 12 West African nations).
Our readers trust us to report truthfully — not just accurately, but justly. That trust is earned daily through verifiable actions: paying consultants on time, citing municipal ordinances alongside academic theory, and refusing to publish until a grandmother in Timor-Leste confirms her grandson’s name is spelled correctly in Tetum orthography. This is not idealism. It is operational discipline — measured in meters, megabytes, minutes, and meticulous cross-referencing.
The numbers matter because people matter. When we write that 24,000 residents of Cusco live in informal settlements lacking piped water (Peru’s INEI 2022 Housing Survey), we do so knowing each digit represents a person waiting for infrastructure. When we describe the 3.2-kilometer walking route from Siem Reap’s Old Market to Angkor Wat’s South Gate — measured precisely using Garmin GPSMAP 66i with GLONASS/Galileo dual-frequency correction — we do so to serve pilgrims, not tourists. Precision is respect. Verification is care. And every comma, citation, and compensation line item is a quiet act of accountability.
This is how journalism serves culture — not as spectacle, but as stewardship.
