ABSTRACT

Recent work on algorithmic bias has shown that understanding the values embedded in technology design processes is important for avoiding social harm. This paper explores the attitudes construed in whitepapers of blockchain technology start-ups. Blockchain technology is a relatively new phenomenon that has informed discourses about the future of governance and economics in relation to the internet. This study aims to understand the values discursively construed in the whitepapers of four blockchain start-ups: Steemit, Creativechain, Democracy Earth, and Bitnation. It adopts a corpus linguistics approach, and uses the Appraisal framework (Martin, J. R., and P. R. R. White. 2005. The Language of Evaluation: Appraisal in English. New York: Palgrave Macmillan) to analyse the evaluative meanings expressed in the whitepaper dataset. This analysis reveals that the blockchain start-ups manifest shared values around the concepts of decentralisation, trust in algorithms, and trust in individuals over institutions. The start-ups enact different political orientations, expressing ideals related to the digital commons, cyber-libertarianism, and capitalism. The corpus-based linguistic analysis used in this study offers a method that may be applicable to other areas of technology discourse where whitepapers and design documents tend to embed covert political and ideological positions.

摘要

最近关于算法偏差的研究表明，了解技术设计过程中嵌入的价值观对于避免社会危害非常重要。本文探讨了区块链技术初创企业白皮书中的态度。区块链技术是一种相对较新的现象，它为有关互联网治理和经济未来的讨论提供了信息。本研究旨在了解四家区块链初创公司（Steemit、Creativechain、Democracy Earth 和 Bitnation）白皮书中论述的价值观。它采用语料库语言学方法，并使用评估框架（Martin、JR 和 PRR White。2005。评估语言：英语评估。纽约：Palgrave Macmillan）分析白皮书数据集中表达的评价意义。这项分析表明，区块链初创企业围绕去中心化、对算法的信任以及对个人而非机构的信任等概念表现出共同的价值观。这些初创企业制定了不同的政治取向，表达了与数字共享、网络自由主义和资本主义相关的理想。本研究中使用的基于语料库的语言分析提供了一种可能适用于白皮书和设计文档往往嵌入隐蔽的政治和意识形态立场的其他技术话语领域的方法。