A Computational Study of Sustainability Reports: How Vagueness Serves Greenwashing
DOI:
https://doi.org/10.6092/issn.2532-8816/21194Keywords:
Natural Language Processing, Sustainability, Greenwashing, Vagueness, Environmental Claims, AIUCD2024Abstract
Today, the automatic processing of large volumes of natural language text is increasingly feasible. However, challenges persist in addressing linguistic ambiguity and vagueness. This study aims to analyze phenomena of vagueness and imprecision in sustainability reports, with a particular focus on greenwashing.
The corpus analyzed consists of 225 Italian-language sustainability reports in PDF format, published by 45 companies between 2017 and 2021. These documents were processed using the Sketch Engine tool and a Python notebook. The analysis began with a targeted search for keywords commonly associated with greenwashing, followed by the annotation of a sample of concordances. Specifically, each extracted example was evaluated to determine whether it met the criteria for an environmental claim and whether it exhibited vagueness. When applicable, instances of vagueness were classified into five semantic categories (quantity, degree, time, category, and softening stance-taking).
The dual contribution of this research lies in its provision of a deeper understanding of linguistic strategies in sustainability reports, promoting corporate transparency and accountability, while simultaneously establishing a foundation for the automated identification of greenwashing-related claims through the development of an Italian-language dataset for Artificial Intelligence training, thereby advancing digital linguistic equality.
References
[1] Bernini, Francesca, e Fabio La Rosa, 2024. «Research in the Greenwashing Field: Concepts, Theories, and Potential Impacts on Economic and Social Value». Journal of Management and Governance 28, fasc. 2 (giugno 2024): 405–44. https://doi.org/10.1007/s10997-023-09686-5.
[2] Bingler, Julia Anna, Mathias Kraus, Markus Leippold, e Nicolas Webersinke, 2024. «How Cheap Talk in Climate Disclosures Relates to Climate Initiatives, Corporate Emissions, and Reputation Risk». Journal of Banking & Finance 164 (luglio 2024): 107191. https://doi.org/10.1016/j.jbankfin.2024.107191.
[3] Channell, Joanna, 1994. Vague Language. Describing English Language. Oxford New York Toronto: Oxford University Press, 1994.
[4] Chen, Minghui, 2018. «Analysis of Vagueness in English Advertisement from the perspective of Adaption Theory.» International Journal of Social Science and Economic Research 3, n. 3 (2018): 1068-1086.
[5] Commissione europea, 2023. «Proposta di direttiva del Parlamento europeo e del Consiglio sull’attestazione e sulla comunicazione delle asserzioni ambientali esplicite (direttiva sulle asserzioni ambientali).» COM/2023/166 final. 22 marzo 2023.
[6] Commissione europea, 2021. «Screening of websites for ‘greenwashing': half of green claims lack evidence.» Press release. Brussels, 28 gennaio 2021.
[7] Cutting, Joan, a c. di. Vague language explored, 2007. Basingstoke [England] ; New York: Palgrave Macmillian, 2007.
[8] De Freitas Netto, Sebastião Vieira, Marcos Felipe Falcão Sobral, Ana Regina Bezerra Ribeiro, e Gleibson Robert Da Luz Soares, 2020. «Concepts and Forms of Greenwashing: A Systematic Review». Environmental Sciences Europe 32, fasc. 1 (dicembre 2020): 19. https://doi.org/10.1186/s12302-020-0300-3.
[9] Hoepner, Andreas, 2024. «Too good to be true: the greenwashers’ box of tricks.» Financial Times. 16 maggio 2024. https://www.ft.com/content/78b3c741-1ab8-48f5-92a8-4e98dfa230ab (consultato il giorno giugno 22, 2024).
[10] Jin, Bixi, 2022. «A corpus-assisted study of vague language in corporate responsibility reports of the cosmetics industry.» Ibérica 43 (2022): 77-102. https://doi.org/10.17398/2340-2784.43.77.
[11] Kilgarriff, Adam, Vít Baisa, Jan Bušta, Miloš Jakubíček, Vojtěch Kovář, Jan Michelfeit, Pavel Rychlý, e Vít Suchomel, 2014. «The Sketch Engine». Lexicography 1, fasc. 1 (1 luglio 2014): 7–36. https://doi.org/10.1007/s40607-014-0009-9.
[12] Lane, Eric L., 2013. «Greenwashing 2.0.» Columbia Journal of Environmental Law 38, n. 2 (2013). https://ssrn.com/abstract=2089078.
[13] Li, Shuangling, 2017 «A Corpus-Based Study of Vague Language in Legislative Texts: Strategic Use of Vague Terms». English for Specific Purposes 45 (gennaio 2017): 98–109. https://doi.org/10.1016/j.esp.2016.10.001.
[14] Lombardi Vallauri, Edoardo, 2022. «Impliciti e persuasione.» Cap. 11 in Pragmatica sperimentale. Manuali Linguistica, a cura di Filippo Domaneschi e Valentina Bambini, 157-170. Bologna: Il Mulino, 2022.
[15] Lombardi Vallauri, Edoardo, Federica Cominetti, e Viviana Masia, 2022. «The Persuasive and Manipulative Power of Implicit Communication». Journal of Pragmatics 197 (agosto 2022): 1–7. https://doi.org/10.1016/j.pragma.2022.04.015.
[16] Mannaioli, Giorgia, 2023. «LA VAGHEZZA LINGUISTICA COME STRATEGIA IMPLICITA PERSUASIVA E FENOMENO DI INTERFACCIA TRA SEMANTICA, SINTASSI E PRAGMATICA. ESEMPI DAL DISCORSO PUBBLICO CONTEMPORANEO IN DIVERSE LINGUE». Italiano LinguaDue 15, fasc. 2 (15 dicembre 2023): 206–21. https://doi.org/10.54103/2037-3597/22028.
[17] Mannaioli, Giorgia, 2025. Vagueness As an Implicitating Persuasive Strategy. 1st ed. Pragmatics and Beyond New Series, v. 350. Amsterdam/Philadelphia: John Benjamins Publishing Company, 2025.
[18] Mannaioli, Giorgia, Alessandro Ansani, Claudia Coppola, e Edoardo Lombardi Vallauri, 2024. «Vagueness as an Implicit-Encoding Persuasive Strategy: An Experimental Approach». Cognitive Processing 25, fasc. 2 (maggio 2024): 205–27. https://doi.org/10.1007/s10339-023-01171-z.
[19] Moodaley, Wayne, e Arnesh Telukdarie, 2023. «A Conceptual Framework for Subdomain Specific Pre-Training of Large Language Models for Green Claim Detection». European Journal of Sustainable Development 12, fasc. 4 (1 ottobre 2023): 319. https://doi.org/10.14207/ejsd.2023.v12n4p319.
[20] Moodaley, Wayne, e Arnesh Telukdarie, 2023. «Greenwashing, Sustainability Reporting, and Artificial Intelligence: A Systematic Literature Review». Sustainability 15, fasc. 2 (12 gennaio 2023): 1481. https://doi.org/10.3390/su15021481.
[21] Moruzzi, Sebastiano, 2012. Vaghezza: confini, cumuli e paradossi. Filosofia / [Laterza]. Roma ; Bari: Laterza, 2012.
[22] Nemes, Noémi, Stephen J. Scanlan, Pete Smith, Tone Smith, Melissa Aronczyk, Stephanie Hill, Simon L. Lewis, A. Wren Montgomery, Francesco N. Tubiello, e Doreen Stabinsky, 2022. «An Integrated Framework to Assess Greenwashing». Sustainability 14, fasc. 8 (8 aprile 2022): 4431. https://doi.org/10.3390/su14084431.
[23] Pinkal, Manfred, 1996. «Vagueness, Ambiguity, and Underspecification». Semantics and Linguistic Theory 6 (15 agosto 1996): 185. https://doi.org/10.3765/salt.v6i0.2775.
[24] Russell, Bertrand, 1923. «Vagueness». Australasian Journal of Psychology and Philosophy 1, fasc. 2 (giugno 1923): 84–92. https://doi.org/10.1080/00048402308540623.
[25] Ruzaitė, Jūratė, 2007. Vague language in educational settings: quantifiers and approximators in British and American English. Europäische Hochschulschriften. Reihe XIV, Angelsächsische Sprache und Literatur ; European university studies. Series XIV, Anglo-Saxon language and literature ; Publications universitaire européennes. Serie XIV, Langue et littérature anglo-saxonnes, Bd. 433 = v. 433 = v. 433. Frankfurt am Main ; New York: Lang, 2007.
[26] Stammbach, Dominik, Nicolas Webersinke, Julia Bingler, Mathias Kraus, e Markus Leippold, 2023. «Environmental Claim Detection». In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers), 1051–66. Toronto, Canada: Association for Computational Linguistics. https://doi.org/10.18653/v1/2023.acl-short.91.
[27] Webersinke, Nicolas, Mathias Kraus, Julia Bingler, e Markus Leippold, 2022. «CLIMATEBERT: A Pretrained Language Model for Climate-Related Text». SSRN Electronic Journal. https://doi.org/10.2139/ssrn.4229146.
[28] Willis, J., 2023. «The Greenwashing Hydra». Planet Tracker. January 2023.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Erica Cutuli

This work is licensed under a Creative Commons Attribution 4.0 International License.