Grammar for Sustainability: GSIC & AI qualitative analysis

Evidano is an AI-powered qualitative data analysis platform that helps researchers analyze interviews, open-ended surveys, and documents. According to the PLoS One study published August 11, 2026, African leaders’ UNGA-2025 English speeches encode sustainability meaning mainly through vocabulary choices and discourse organization, offering a ready corpus for grammar-based Education for Sustainable Development (ESD). This post explains the study’s key numbers and pedagogical recommendation, shows how AI-enabled qualitative research speeds each GSIC phase, and gives practical next steps for language teachers and researchers.

Key Takeaways

According to the PLOS One study published August 11, 2026, a 50, 644-token corpus of 21 African leaders’ UNGA-2025 speeches shows that sustainability meaning is built predominantly through lexical choice (48.6%) and discourse organization (26.5%); the study’s pedagogy is the Grammar-for-Sustainability Instructional Cycle (GSIC) (PLOS One).

The corpus comprised 50, 644 tokens from 21 English-language speeches delivered during the 80th UNGA session in September–October 2025, and contained 1, 956 sentences and 1, 515 paragraphs, according to Lasekan et al., PLoS One (August 11, 2026).
Quantitatively, lexical resources accounted for 48.6% of sustainability expression and discourse organization 26.5%, with pragmatic strategies contributing 16.5% (Lasekan et al., PLoS One, 2026).
Syntactic markers mapped to competencies: over 72% of modal verbs, 75.6% of passive constructions, and 72.4% of conditional forms occurred in SDG contexts in the corpus (Lasekan et al., PLoS One, 2026).
The study proposes GSIC, a five-phase cycle (noticing, analysis, transformation, production, reflection) to teach both grammar and UNESCO sustainability competencies, and provides replication materials in its Supplementary Information (Lasekan et al., PLoS One, 2026).

What happened and how the analysis worked

What happened: researchers compiled and analyzed a 50, 644-token corpus of 21 African leaders’ English UNGA-2025 speeches using corpus tools (AntConc) and Python (spaCy) with mixed quantitative–qualitative coding, as reported by Lasekan et al., PLoS One (August 11, 2026).

According to the methods reported in PLoS One (Lasekan et al., 2026), sentences were tagged as sustainability discourse only if they contained at least one item from a 327-item SDG lexicon derived from official UN SDG documentation.

According to Lasekan et al., PLoS One (2026), linguistic analysis was organized across five levels (lexical, morphological, syntactic, pragmatic, and discourse) with quantitative normalization per 10, 000 tokens and manual validation (inter-rater Cohen’s kappa: argument stage κ =.81; pragmatic function κ =.85; SDG relevance κ =.92).

Because the study made its cleaned corpus and replication files available in Supplementary Materials, researchers can reproduce SDG tagging and feature extraction using the provided CSV/JSON outputs and the Methods Summary (Lasekan et al., PLoS One, 2026).

Findings snapshot

Date / Source	Metric	Value	Implication
August 11, 2026 / PLOS One	Corpus size	50, 644 tokens; 21 speeches; 1, 956 sentences	Authentic, high-stakes diplomatic input suitable for advanced EFL GSIC tasks (Lasekan et al., PLoS One, 2026).
August 11, 2026 / PLOS One	Lexical vs discourse share	Lexical 48.6%; Discourse 26.5%; Pragmatic 16.5%	Teach vocabulary and problem–solution sequencing first, then pragmatics and syntax (Lasekan et al., PLoS One, 2026).
August 11, 2026 / PLOS One	Syntactic SDG alignment	Modal verbs >72% in SDG contexts; Passive 75.6%; Conditionals 72.4%	Use concordance-driven tasks on modals, passives, and if–then clauses to build normative and systems-thinking competence (Lasekan et al., PLoS One, 2026).
August 11, 2026 / PLOS One	Competency emphasis	Strategic 42.6%; Interpersonal 26.4%	Prioritize problem–solution and collective-pronoun tasks to model coalition and multilateral framing (Lasekan et al., PLoS One, 2026).

Implications for language teachers and qualitative researchers

Language teachers can operationalize GSIC immediately by using UNGA-2025 concordance lines to teach modal obligation, conditional reasoning, and collective pronouns, because Lasekan et al. (PLOS One, August 11, 2026) show these features map to UNESCO sustainability competencies.

According to the PLoS One analysis (Lasekan et al., 2026), prioritizing lexical and discourse-level tasks helps students build meaningful SDG arguments because vocabulary carried 48.6% of sustainability meaning and problem–solution staging was dominant.

Qualitative researchers can replicate the approach: the PLoS One team provides the full SDG lexicon (327 items), the cleaned corpus, and concordance exports in Supplementary Files to support reproducible mixed-methods analysis (Lasekan et al., PLoS One, 2026).

For classroom feasibility, Lasekan et al. (PLOS One, 2026) recommend GSIC for upper-intermediate to advanced EFL learners and note the cycle can be implemented over two to four sessions or adapted for low-resource classrooms with printed concordance lines.

How Evidano helps: AI-enabled qualitative workflows for GSIC

Problem: compiling and tagging a corpus is time-consuming

Solution: Evidano automates ingestion of documents and web scraping of public speeches, speeding corpus compilation from hours to minutes and preserving original source metadata (Evidano features).

According to the PLoS One replication materials (Lasekan et al., PLoS One, 2026), reproducible tokenization and lemmatization are essential; Evidano exports standardized CSV/JSON that match those formats to ease replication.

Problem: manual concordance and multi-level coding is labor-intensive

Solution: Evidano performs thematic, frequency, and co-occurrence analyses automatically and produces concordance displays and exportable codebooks for modal verbs, passives, and SDG lexicon matches, enabling teachers to create GSIC noticing and analysis activities in minutes (Evidano features).

Solution detail: Evidano’s AI chat over your documents lets educators ask natural-language questions like “show me all sentences with must + climate” and get instant concordance lines ready for Phase 1 of GSIC.

Problem: scaffolding production and cross-segment comparison is hard

Solution: Evidano generates cross-segment analyses (e.g., by country, SDG theme, or speaker) and visualizations (word clouds, co-occurrence networks, hierarchical codes) that teachers can use for Phase 4 production prompts and peer review rubrics.

Security note: Evidano uses proprietary LLMs tuned for qualitative research and encrypts data; customer data is not used to train third-party models, supporting classroom privacy and research integrity.

FAQ: grammar for sustainability

What is 'grammar for sustainability' and why does it matter?

Answer: Grammar for sustainability is the idea that specific grammatical forms (modals, conditionals, passives, pronouns, discourse sequencing) enact sustainability competencies in communication.

Support: According to Lasekan et al. (PLOS One, August 11, 2026), modal verbs such as must and should encode normative competence while conditional structures support systems-thinking, so teaching these forms simultaneously develops linguistic control and sustainability reasoning.

How can I reproduce the PLoS One UNGA-2025 analysis for a local corpus?

Answer: Reproduce the study by building a lemmatized corpus, applying an SDG lexicon, extracting concordances, and mapping features to competencies using the study’s Methods Summary and Supplementary Files.

Support: Lasekan et al. (PLOS One, 2026) provide a 327-item SDG lexicon, token-processing documentation, and CSV/JSON exports in Supplementary Materials to enable replication and adaptation.

Can GSIC work in low-resource classrooms?

Answer: Yes, GSIC can be implemented with printed concordance lines and teacher-curated excerpts when digital tools are unavailable.

Support: The authors note GSIC is adaptable and recommend teacher training in corpus-informed pedagogy; they state the cycle can be run over two to four sessions or adapted for paper-based materials (Lasekan et al., PLOS One, 2026).

How can AI help teachers implement GSIC quickly?

Answer: AI-enabled qualitative tools speed corpus assembly, SDG tagging, concordance generation, and cross-segment comparison so teachers can focus on task design and feedback.

Support: Using an AI platform, teachers can produce targeted Phase 1 noticing materials and Phase 4 production scaffolds in minutes rather than days; Evidano supports these steps with document ingestion, concordance, and AI chat features (Evidano features).

Conclusion & Next Steps

According to Lasekan et al. (PLOS One, August 11, 2026), the UNGA-2025 corpus demonstrates that grammar encodes sustainability competencies and that GSIC offers a practical five-phase classroom cycle to teach both language and ESD skills.

Teachers and researchers can reproduce the study’s methods using the provided Supplementary Materials and accelerate implementation with AI-enabled qualitative platforms that automate corpus tasks and produce concordance-ready outputs.

If you want to pilot GSIC in a class or research project, start by downloading the PLoS One replication files and using an AI qualitative tool to extract modal and conditional concordances; see the study here: PLOS One.

Ready to prototype GSIC with automated corpus workflows? Try Evidano for free to ingest speeches, generate SDG concordances, and build classroom-ready tasks in minutes.

Topics

grammar for sustainability
GSIC pedagogy
corpus-informed grammar instruction
AI qualitative analysis

Keep reading

Browse all articles