Anonymity promotes free speech by protecting the identity of people who might otherwise face negative consequences for expressing their ideas. Wrongdoers, however, often abuse this invisibility cloak. Defenders of anonymity online emphasise its value in advancing public debate and safeguarding political dissension. Critics emphasise the need for identifiability in order to achieve accountability for wrongdoers such as trolls. The problematic tension between anonymity and identifiability online lies in the desirability of having low costs (no repercussions) for desirable (...) speech and high costs (appropriate repercussions) for undesirable speech. If we practice either full anonymity or identifiability, we end up having either low or high costs in all online contexts and for all kinds of speech. I argue that free speech is compatible with instituting costs in the form of repercussions and penalties for controversial and unacceptable speech. Costs can minimise the risks of anonymity by providing a reasonable degree of accountability. Pseudonymity is a tool that can help us regulate those costs while furthering free speech. This article argues that, in order to redesign the Internet to better serve free speech, we should shape much of it to resemble an online masquerade. (shrink)

This book aims to understand why women are the targets of online hate speech and how we can stop this from occurring.

In this paper we explore quarantining as a more ethical method for delimiting the spread of Hate Speech via online social media platforms. Currently, companies like Facebook, Twitter, and Google generally respondreactivelyto such material: offensive messages that have already been posted are reviewed by human moderators if complaints from users are received. The offensive posts are onlysubsequentlyremoved if the complaints are upheld; therefore, they still cause the recipients psychological harm. In addition, this approach has frequently been criticised for (...) delimiting freedom of expression, since it requires the service providers to elaborate and implement censorship regimes. In the last few years, an emerging generation of automatic Hate Speech detection systems has started to offer new strategies for dealing with this particular kind of offensive online material. Anticipating the future efficacy of such systems, the present article advocates an approach to online Hate Speech detection that is analogous to the quarantining of malicious computer software. If a given post is automatically classified as being harmful in a reliable manner, then it can be temporarily quarantined, and the direct recipients can receive an alert, which protects them from the harmful content in the first instance. The quarantining framework is an example of more ethical online safety technology that can be extended to the handling of Hate Speech. Crucially, it provides flexible options for obtaining a more justifiable balance between freedom of expression and appropriate censorship. (shrink)

The concept of counterspeech denotes a non-coercive and non-censoring method for reacting to harmful speech, with the aim of impeding or at least diminishing its damaging effects. Remarkable work is being done by researchers and activist groups on elaborating practical strategies of countering hate speech online. Though, research in moral and political philosophy exploring the effectivity of counterspeech and grounding the reasons for engaging in it still remains in its early stages. In the following paragraphs I will (...) address recent contributions which elaborate on the viability and normative aspects of counterspeech. Outlining their valuable insights, but also their failure to give due importance to the peculiarities of online speech dynamics, I will highlight relevant features on which future research about online harmful speech and counterspeech can build. (shrink)

A lot of subordinating speech has moved online, which raises several questions for philosophers. Can current accounts of oppressive speech adequately capture digital hate? How does the anonymity of online harassers contribute to the force of their speech? This paper examines online abuse and argues that standard accounts of licensing and accommodation are not up to the task of explaining the authority of online hate speech, as speaker authority often depends on the (...) community in more ways than these accounts suggests. Instead I argue that online abusive speech is best understood as collective subordinating speech acts, as their authority is drawn from an ad hoc collective. I argue that anonymity and shared language offer online abusers a path to a type of group-authority that explains the harm their speech is capable of. I close by suggesting that similar considerations are in play for IRL subordinating speech, and that online abuse helps to reveal key features of subordinating speech across mediums that are under-emphasized in the existing literature. (shrink)

Government agencies in New Zealand are not required to systematically collect data on online hate speech, thus, there is a lack of longitudinal evidence regarding this phenomenon. This report presents trends in personal experiences of and exposure to online hate speech among adult New Zealanders based on nationally representative data. The findings from this study are also compared with results from a similar research study conducted in 2018. In addition, this report explores people’s perceptions about other (...) issues related to hate speech. The goal of this study is to fill some knowledge gaps regarding the extent and nature of hate speech in New Zealand. It also seeks to provide reliable and robust evidence to inform public conversation on the topic and the implementation of strategies and activities to ensure all New Zealanders, no matter what their characteristics, are safe online to benefit from the multiple opportunities of the digital environment. (shrink)

Government agencies in New Zealand are not required to systematically collect data on online hate speech, thus, there is a lack of longitudinal evidence regarding this phenomenon. This report presents trends in personal experiences of and exposure to online hate speech among adult New Zealanders based on nationally representative data. The findings from this study are also compared with results from a similar research study conducted in 2018. In addition, this report explores people’s perceptions about other (...) issues related to hate speech. The goal of this study is to fill some knowledge gaps regarding the extent and nature of hate speech in New Zealand. It also seeks to provide reliable and robust evidence to inform public conversation on the topic and the implementation of strategies and activities to ensure all New Zealanders, no matter what their characteristics, are safe online to benefit from the multiple opportunities of the digital environment. (shrink)

The present paper aims to report on the preliminary findings from the initial stages of ongoing research on hate speech in Lithuanian online comments. Comments are marked strongly by such phenomena as flaming and trolling; therefore, in this genre we can expect a high degree of hostility, obscenity, high incidence of insults and aggressive lexis, which can inflict harm to individuals or organizations. The goal of the current research is thus to make an attempt to identify some features (...) of verbal aggression in Lithuanian by applying the principles and instruments of corpus linguistics, which proved to be a useful approach when dealing with such issues as trolling. It is expected that further analysis of those features will help to identify and define formal linguistic criteria that could facilitate identification of hate speech in public discourse. The data has been obtained from the Lithuanian corpus of user-generated comments collected from one major Lithuanian portal, www.delfi.lt. The corpus consists of all the comments posted in the year 2014 and in total includes 17,909 comments, which make up 1,160,109 words. For the initial data analysis, linguistic aspects, such as wordlists, collocations, and formulaic language, were analysed by using the AntConc software. The interpretations of the results are still very tentative, but what the initial findings show is that overt aggression does not feature among the most frequent and most salient features of comments. Aggression is, in our data, indirectly expressed through creative language use, which can mainly be studied through qualitative analysis. (shrink)

We provide a new text corpus from the social medium Telegram, which is rich in indirect forms of divisive speech. We scraped all messages from one channel of supporters of Donald Trump, covering a large part of his presidency from late 2016 until January 2021. The discussion among the group members over this long time period includes the spread of disinformation, disparaging of out-group members, and other forms of offensive speech. To encourage research into such practices of poisoning (...) public political discourse, we added automatic annotations of offensive language to all messages. We further added manual annotations of harmful language to a portion of the posts in order to enable the analysis of more implicit forms of online harm. (shrink)

This corpus analysis of linguistic and semantic features in French parliamentary debates concerning online hate speech regulation, highlights tensions between state powers and private rights. Two key themes are identified: first, the _problem of definition_: how such online content is defined in the debates, and second, the _problem of regulation_: how the debates negotiate the supra-jurisdictional and individual jurisdiction issues involved, in regulating both the global online content and the responsibilities of the owners of the platforms (...) who manage the content. For this analysis, the authors created a corpus of French National Assembly Examination Committee discussions and public sessions between July 2019 to January 2020 discussing the _Loi Avia _(Avia Law). The corpus was analysed using quali-quantitative methods. Overall, the study combines a top-down (global) and a bottom-up (local) analysis of the data, applying corpus semantic approaches (Lecolle et al. in Langages 210(2):35–54, 2018; Rastier in The Routledge handbook of semantics, Routledge, pp 507–522, 2015) to contextualize the analysis of the meta-discourse (Hyland in Meta-discourse: exploring interaction in writing, 1st ed. Bloomsbury Academic, Bloomsbury Collections, 2018. https://doi.org/10.5040/9781350063617 ). The analysis highlights the linguistic features revealing the power dynamics inherent within the discourse and counter-discourse of the dominant and opposition political parties respectively. (shrink)

The research study aimed at assessing the various difficulties that speech teachers had in delivering lessons during the distance learning era. The various phenomena and themes that emerged from the survey were determined using a descriptive research design. The survey was conducted to collect information about the various characteristics and behavior of speech teachers. It also explored the factors that influenced their decisions and actions when it came to addressing the pandemic. The repeated themes that emerged, include low (...) motivation, altered activities, unstable internet connection, comprehension, questionable output integrity, delayed submissions, in-person monitoring, and the system of communicating with inactive students in online mode. (shrink)

Jerkish speech on online platforms is at risk of being overlooked as a result of being comparatively insignificant next to the existence of explicit hate speech or other online harms. In this paper I approach online jerkish speech from a republican perspective. I discuss two ways in which republicans can account for jerkish speech. First, jerkish speech could amount to micro-domination, referring to instances of domination that are relatively inconsequential by themselves but (...) problematic when considered in aggregate. Second, jerkish speech can be considered an instantiation of systemic domination, where groups and members are structurally subordinated. I briefly discuss the role of regulatory bans on jerkish speech, before suggesting that Suzanne Whitten’s proposal for critical civility—as an appropriate answer to jerkish speech in online environments—might be too demanding. This might be alleviated by leveraging the role of online platforms. (shrink)

-/- Hate speech is a concept that many people find intuitively easy to grasp, while at the same time many others deny it is even a coherent concept. A majority of developed, democratic nations have enacted hate speech legislation—with the contemporary United States being a notable outlier—and so implicitly maintain that it is coherent, and that its conceptual lines can be drawn distinctly enough. Nonetheless, the concept of hate speech does indeed raise many difficult questions: What does (...) the ‘hate’ in hate speech refer to? Can hate speech be directed at dominant groups, or is it by definition targeted at oppressed or marginalized communities? Is hate speech always ‘speech’? What is the harm or harms of hate speech? And, perhaps most challenging of all, what can or should be done to counteract hate speech? -/- In part because of these complexities, hate speech has spawned a vast and interdisciplinary literature. Legal scholars, philosophers, sociologists, anthropologists, political theorists, historians, and other academics have each approached the topic with exceeding interest. In this current article, however, we cannot hope to cover how these many disciplines have engaged with the concept of hate speech. Here, we will focus most explicitly on how hate speech has been taken up within philosophy, with particular emphasis on issues such as: how to define hate speech; what are the plausible harms of hate speech; how an account of hate speech might include both overt expressions of hate (e.g., the vitriolic use of slurs) as well as more covert, implicit utterances (e.g., dogwhistles); the relationship between hate speech and silencing; and what might we do to counteract hate speech. (shrink)

This dissertation fits within the literature on subordinating speech and aims to demonstrate that how language subordinates is more complex than has been described by most philosophers. I argue that the harms that subordinating speech inflicts on its targets (chapter one), the type of authority that is exercised by subordinating speakers (chapters two and three), and the expansive variety of subordinating speech acts themselves (chapter three) are all under-developed subjects in need of further refinement—and, in some cases, (...) large paradigm shifts. I also examine cases that have yet to be adequately addressed by philosophers working on this topic, like the explosion of abusive speech online (chapter four) or the distinctive speech acts of protest groups (chapter five). I argue that by considering these alongside the ‘paradigm’ cases of subordinating speech that inform most models, we are better able to capture the lived realities of this phenomena, as described by members of groups targeted by such speech. I develop a novel account of speaker authority to explain the variety of pragmatic effects subordinating speech generates. Instead of seeing this authority as reducible to either a formal position or a merely local, linguistic phenomenon, I argue for a conception of speaker authority that is a richly contextual social fact, distributed unevenly among members of different social groups. I also develop an account of collective authority that explains how a group of speakers can join together to subordinate in a way that no individual speaker is capable of doing. This account, I argue, is better able to explain the social reality of subordinating speech than individualist models. Overall, I show how a more fine-grained account of subordinating speaker authority gives us a more accurate picture of the different subordinating speech acts available to different speakers, along with how these may harm their targets. (shrink)

White lies, lies that are beneficial to the addressee and uttered out of concern for them, are pervasive in daily life. This raises a prima facie question: if white lies are so common, are they still lies? I tackle this question from the perspective of speech act theory, using the case of white lies to shed light on the broader question of whether lying itself can be considered a type of speech act. Adopting Sbisà’s (Uptake and Conventionality in (...) Illocution. Lodz Papers in Pragmatics 5 33–52. Reprinted in Marina Sbisà (forthcoming), Essays on Speech Acts and Other Topics in Pragmatics. Oxford: Oxford University Press, 2009) neo-Austinian approach that places the emphasis on the achievement of certain conventional effects, I claim that malicious lying cannot be a type of speech act. However, white lies are a type of speech act, a claim which finds support in three types of evidence: the conventionality of means by which white lies are performed, their online processing, and their acquisition. Moreover, the existence of a continuum between white (harmless) lies (which are a type of speech act) and malicious (harmful) lies (which are not) suggests that whether an utterance performs a certain type of speech act or not can be a matter of degree. In other words, the property itself of being a speech act of a certain kind, an utterance’s speech-act-hood, so to speak, may be graded. This is an interesting result, which applies to speech acts more generally, beyond lying. (shrink)

Building on the work of Peter Hinst and Geo Siegwart, we develop a pragmatised natural deduction calculus, i.e. a natural deduction calculus that incorporates illocutionary operators at the formal level, and prove its adequacy. In contrast to other linear calculi of natural deduction, derivations in this calculus are sequences of object-language sentences which do not require graphical or other means of commentary in order to keep track of assumptions or to indicate subproofs. (Translation of our German paper "Ein Redehandlungskalkül. Ein (...) pragmatisierter Kalkül des natürlichen Schließens nebst Metatheorie"; online available at http://philpapers.org/rec/CORERE.). (shrink)

We explore the speech act of amplification and its newfound prominence in online speech environments. Then we point to some puzzles this raises for the strategy of ‘fighting speech with more speech’.

Harmful and dangerous language is frequent in social media, in particular in spaces which are considered anonymous and/or allow free participation. In this paper, we analyse the language in a Telegram channel populated by followers of Donald Trump, in order to identify the ways in which harmful language is used to create a specific narrative in a group of mostly like-minded discussants. Our research has several aims. First, we create an extended taxonomy of potentially harmful language that includes not only (...) hate speech and direct insults, but also more indirect ways of poisoning online discourse, such as divisive speech and the glorification of violence. We apply this taxonomy to a large portion of the corpus. Our data gives empirical evidence for harmful speech such as in/out-group divisive language and the use of codes within certain communities which have not often been investigated before. Second, we compare our manual annotations to several automatic methods of classifying hate speech and offensive language, namely list based and machine learning based approaches. We find that the Telegram data set still poses particular challenges for these automatic methods. Finally, we argue for the value of studying such naturally occurring, coherent data sets for research on online harm and how to address it in linguistics and philosophy. (shrink)

Using speech act theory from the Philosophy of Language, this paper attempts to develop an ethical framework for the phenomenon of speech processing. We use the concepts of the illocutionary force and the illocutionary content of a speech act to explain the ethics of speech processing. By emphasizing the different stages involved in speech processing, we explore the distinct ethical issues that arise in relation to each stage. Input, processing, and output are the different ethically (...) relevant stages under which a spoken item or a speech navigates within the range of speech-processing modules. Employing the illocutionary force-content distinction, we specify and characterize the input-related ethical issues, the output-related ethical issues, and the processing-related ethical issues involved in speech processing. Together with illocutionary force-content distinction, we employ the data-information distinction to characterize the stage-wise ethical issues in the phenomenon of speech processing as the ethics of collecting (speech) data, the ethics of contextualizing (speech) data/information, and the ethics of releasing the con-textualized information (processed speech). Immediate ethical issues that arise from the range of speech processing modules are distinguished from distant ethical issues. We also indicate the nature of ethical issues that arise from Speaker Independent speech technologies. (shrink)

This study delves into the manifestation and characteristics of hate speech in the Greek online public sphere, specifically exploring its most prominent forms, namely racism, anti-immigrant sentiment, nationalism, sexism, and homophobia/transphobia. Combining quantitative and qualitative methods, the research analyzes popular Greek online news media. It aims to uncover the visibility and operational patterns of hate speech, addressing key questions about its prevalence and presentation on these platforms. Findings reveal the normalization of discriminatory speech, particularly sexism (...) and nationalism, in the digital landscape. Moreover, distinct themes and linguistic elements associated with different forms of hate speech highlight the intertwined nature of these discriminatory expressions. This research sheds light on the pervasive nature of hate speech in Greece’s online realm, emphasizing its implications for democratic principles. (shrink)

Speech acts are a staple of everyday communicative life, but only became a topic of sustained investigation, at least in the English-speaking world, in the middle of the Twentieth Century.[1] Since that time “speech act theory” has been influential not only within philosophy, but also in linguistics, psychology, legal theory, artificial intelligence, literary theory and many other scholarly disciplines.[2] Recognition of the importance of speech acts has illuminated the ability of language to do other things than describe (...) reality. In the process the boundaries among the philosophy of language, the philosophy of action, the philosophy of mind and even ethics have become less sharp. In addition, an appreciation of speech acts has helped lay bare an implicit normative structure within linguistic practice, including even that part of this practice concerned with describing reality. Much recent research aims at an accurate characterization of this normative structure underlying linguistic practice. (shrink)

Despite its long history of investigating sociality, phenomenology has, to date, said little about online sociality. The phenomenological tradition typically claims that empathy is the fundamental way in which we experience others and their experiences. While empathy is discussed almost exclusively in the context of face-to-face interaction, I claim that we can empathetically perceive others and their experiences in certain online situations. Drawing upon the phenomenological distinction between the physical, objective body and the expressive, lived body, I: (i) (...) highlight that empathy involves perceiving the other’s expressive, lived body, (ii) show that the lived body is not tied to the physical body and that empathy can take place outside of face-to-face interactions, and (iii) argue that the lived body can enter online space and is empathetically available to others there. I explore two ways in which the other’s lived body enters online space and can be empathetically perceived: first, in cases where our face-to-face encounter is technologically-mediated over video link and, second, by showing how the other’s texts, as speech, can form part of the other’s lived body. Investigating empathy online not only furthers our understanding of online encounters but also leads to a refined conception of empathy more generally. (shrink)

I have argued elsewhere that in past history, freedom of speech, whether granted to few or many, was granted as bestowing some important benefit. John Stuart Mill, for example, in On Liberty, saw it as enabling us to learn from each other through discussion. By the test of benefit, I here argue that social media that are funded through trade in our personal data with advertisers, including propagandists, cannot claim to be supporting free speech. We lose our freedoms, (...) if the personal data we entrust to online social media are used to target us with information, or disinformation, tailored as persuasive to different personalities, in order to maximize revenue from advertisers or propagandists. Among the serious consequences described, particularly dangerous because of its effect on democracy, is the use of such targeted advertisements to swing voting campaigns. Control is needed both of the social media and of any political parties that pay social media for differential targeting of voters based on personality. Using UK government documents, I recommend legislation for reform and enforcement. (shrink)

The Internet presents opportunities for corporations to efficiently build their brands online and to enhance their global reach. But there are threats as well as opportunities, since anti-branding and free-riding activities are easier in cyberspace. One such threat is theunauthorized incorporation of a trademark into a domain name. This can lead to trademark dilution and cause consumer confusion. But some users claim a right to use these trademarks for the purpose of parody or criticism. Underlying these trademark conflicts is (...) the familiar tension between property rights and free speech rights. While some trademark scholars are reluctant to consider a trademark as property, we find strong support for the property paradigm in Hegel’s philosophy. Assuming that a trademark is an earned property right, we propose that a trademark owner should be allowed to control the permutations of its trademark incorporated into domain names unless a reasonable person would not confuse that domain name with the company’s mark. But we also conclude that there must be latitude to employ a domain name for negative editorial comment, so long as the source and purpose of that domain name is plainly apparent. (shrink)

The Internet presents opportunities for corporations to efficiently build their brands online and to enhance their global reach. But there are threats as well as opportunities, since anti-branding and free-riding activities are easier in cyberspace. One such threat is theunauthorized incorporation of a trademark into a domain name. This can lead to trademark dilution and cause consumer confusion. But some users claim a right to use these trademarks for the purpose of parody or criticism. Underlying these trademark conflicts is (...) the familiar tension between property rights and free speech rights. While some trademark scholars are reluctant to consider a trademark as property, we find strong support for the property paradigm in Hegel’s philosophy. Assuming that a trademark is an earned property right, we propose that a trademark owner should be allowed to control the permutations of its trademark incorporated into domain names unless a reasonable person would not confuse that domain name with the company’s mark. But we also conclude that there must be latitude to employ a domain name for negative editorial comment, so long as the source and purpose of that domain name is plainly apparent. (shrink)

Inner speech is known as the “little voice in the head” or “thinking in words.” It attracts philosophical attention in part because it is a phenomenon where several topics of perennial interest intersect: language, consciousness, thought, imagery, communication, imagination, and self-knowledge all appear to connect in some way or other to the little voice in the head. Specific questions about inner speech that have exercised philosophers include its similarities to, and differences from, outer speech; its relationship to (...) reasoning and conceptual thought; its broader cognitive roles—especially within metacognition and self-knowledge; and the role it can play in explanations of auditory verbal hallucinations and “thought insertion”. (shrink)

The category of expressive speech acts has traditionally proven elusive of definition in contrast to other types of speech acts. This might explain why this group of speech acts has been less researched. The present paper aims to redress this imbalance by analysing the expressive speech acts performed by two groups of university students in two educational chats, carried out in English or in Spanish, respectively. The main purpose of the study is to find out if (...) students express their emotions when interacting online and, if the use of their mother tongue or not affects their performance of expressive speech acts in terms of frequency and type. To this purpose, Weigand’s taxonomy of speech acts was followed, since it provides a more systematic delimitation of the traditional category of “expressive acts”. Her distinction between emotives and declaratives was thus applied to the dataset under scrutiny. Results show that students opt for performing declarative acts but refrain from expressing their own emotions in an educational setting. As for the use of English or Spanish, no significant differences were observed, which reveals that the use of their mother tongue does not seem to affect the kind of acts performed. (shrink)

The Internet presents opportunities for corporations to efficiently build their brands online and to enhance their global reach. But there are threats as well as opportunities, since anti-branding and free-riding activities are easier in cyberspace. One such threat is theunauthorized incorporation of a trademark into a domain name. This can lead to trademark dilution and cause consumer confusion. But some users claim a right to use these trademarks for the purpose of parody or criticism. Underlying these trademark conflicts is (...) the familiar tension between property rights and free speech rights. While some trademark scholars are reluctant to consider a trademark as property, we find strong support for the property paradigm in Hegel’s philosophy. Assuming that a trademark is an earned property right, we propose that a trademark owner should be allowed to control the permutations of its trademark incorporated into domain names unless a reasonable person would not confuse that domain name with the company’s mark. But we also conclude that there must be latitude to employ a domain name for negative editorial comment, so long as the source and purpose of that domain name is plainly apparent. (shrink)

In this contribution, I will argue that the image of a balance is often used to defend the idea of a trade-off. To understand the drawbacks of this line of thought, I will explore the relationship between online security technologies and fundamental rights, notably privacy, nondiscrimination, freedom of speech and due process. After discriminating between three types of online security technologies, I will trace the reconfiguration of the notion of privacy in the era of smart environments. This (...) will lead to an inquiry into the metaphor of the scale, building on the triple test regarding the justification of the limitation of fundamental rights such as privacy. The conclusion will be that in the case of a trade-off, infringing measures will have to be balanced by effective safeguards. No trade-off without balance. (shrink)

This paper investigates whether search engines and other new modes of online communication should be covered by free speech principles. It criticizes the analogical reason-ing that contemporary American courts and scholars have used to liken search engines to newspapers, and to extend free speech coverage to them based on that likeness. There are dissimilarities between search engines and newspapers that undermine the key analogy, and also rival analogies that can be drawn which don’t recommend free speech (...) protection for search engines. Partly on these bases, we argue that an analogical approach to questions of free speech coverage is of limited use in this context. Credible verdicts about how free speech principles should apply to new modes of online com-munication require us to re-excavate the normative foundations of free speech. This method for deciding free speech coverage suggests that only a subset of search engine outputs and similar online communication should receive special protection against government regulation. (shrink)

According to a view widely held in the media and in public discourse more generally, online hating is a social problem on a global scale. However, thus far there has been little scientific literature on the subject, and, to our best knowledge, there is even no established scholarly definition of online hating and online haters in the first place. The purpose of this manuscript is to provide a new perspective on online hating by, first, distinguishing (...) class='Hi'>online hating from the phenomena it is often confused with, such as trolling, cyberstalking, and online hate speech, and, second, by proposing an operational definition of online hating and online haters based on ethnographic interviews and surveys of the existing scholarly literature. (shrink)

Algorithm engineering is sometimes portrayed as a new 21st century return of manipulative social engineering. Yet algorithms are necessary tools for individuals to navigate online platforms. Algorithms are like a sensory apparatus through which we perceive online platforms: this is also why individuals can be subtly but pervasively manipulated by biased algorithms. How can we better understand the nature of algorithm engineering and its proper function? In this chapter I argue that algorithm engineering can be best conceptualized as (...) a type of environmental engineering aimed at making the online environment more hospitable to human use, in particular by safeguarding the conditions that allow for trust. (shrink)

Purpose The Hon’ble Supreme Court of India, in a landmark judgment, scrapped a draconian law [Section 66 ] that gave the police absolute power to put behind bars anybody who was found posting offensive or annoying comments online. This paper aims to examine the take of people on the “Free Speech via Social Media” issue and their attitude towards the way sensitive messages/information are posted, shared and forwarded on social media, especially, Facebook. Design/methodology/approach The research was carried out (...) on a sample of 200 social media users, all picked up randomly, from five Indian states/Union Territories. Data were collected through a questionnaire, and users were contacted through e-mail. Data collected were analyzed through the Kolmogorov–Smirnov Z test. Findings The findings indicate that hate posts/messages are on the rise, and more and more users are joining in. Besides, prosecution happens only when the aggrieved party is influential or powerful. Practical implications The findings of this research give a strong insight into the social media behaviour of users in relation to hate contents/posts. The study establishes the fact that Indian people are in favour of free speech, but with a sense of restraint and responsibility. The work could form the basis for future research on various aspects of hate speech on social media. Researchers could study the trials and prosecutions that have happened over the past few years and whether punishment has acted as a deterrent. Originality/value The research is likely to be important for those involved in work on freedom of speech or hate speech through social media. Social networking sites such as Facebook would also get some insights into users’ perception towards free and hate speech mechanism on social media. (shrink)

According to many theories of testimony, acts of testimony confer certain epistemic rights upon recipients, e.g., the right for the recipient to complain or otherwise hold the testifier responsible should the content of that testimony turn out to be false, and the right to “pass the epistemic buck”, such that the recipient can redirect relevant challenges they may encounter back to the testifier. While these discussions do not explicitly exclude testimonial acts that occur online, they do not specifically address (...) them, either. Here, then, I will ask the following questions: do the differences between communicating in online and offline spaces affect our testimonial epistemic rights, and if so, how? While there is no singular “online space”, here I will focus on such spaces in which users communicate with one another, and in which communicated information can be vetted by other users (for example, social media). I argue that the characteristics of online testimony should make us think about testimonial epistemic rights differently, in two ways. First, whereas such rights have traditionally been conceived of as existing between the recipient and testifier, in many different types of online communication these rights exist between the recipient and a community. This is a result of the fact that online testimony is mediated, and in some cases partially determined by, a community of users. As such, testimonial epistemic rights in online spaces may be widely extended: while the original testifier still bears the brunt of responsibility for challenges, and is the primary buck-passee, all other members of the relevant community will also bear some such responsibilities. Second, the grounds of testimonial epistemic rights may differ in online spaces. Existing theories tend to ground such rights either in assurances provided by the testifier, or else norms that govern speech acts. I argue that testimony in online spaces should cause us to look to a third option, what I call norms of information sharing. The idea is that, given the highly social nature of online communication, a recipient acquires testimonial epistemic rights in virtue of having a reasonable expectation that information that is shared and vetted by the community meets certain standards. The grounds of online testimonial epistemic rights, then, is not primarily interpersonal or norms-based, but social. (shrink)

The unique role of the Internet in today’s society, and the extensive reach and potentially profound impact of much Internet content, raise philosophically interesting and practically urgent questions about the responsibilities of various agents, including individual Internet users, governments, and corporations. Raphael Cohen-Almagor’s Confronting the Internet’s Dark Side is an extremely valuable contribution to the emerging discussion of these important issues. In this paper, I focus on the obligations of Internet Service Providers and Web Hosting Services with respect to (...) class='Hi'>online hate speech. I argue that although Cohen-Almagor is correct that we should not understand these companies’ obligation to protect and promote freedom of expression as always taking priority over potentially competing values, his argument that they ought to deny service to those who would engage in hate speech online is less than fully convincing. In part this is because his definition of hate speech is, in important respects, overly broad. In addition, I argue that the analogies to which he appeals in defense of his position are flawed, and that a more accurate analogy appears to provide at least some support to the view that ISP’s and WHS’s ought not deny service to those who would engage in at least some forms of hate speech. (shrink)

Our time is marked by a resurgence of hate that threatens to increase oppression. Social media has contributed to this by acting as a medium through which hate speech is spread. How should we model the spread of hate? This article considers two models. First, I consider a simple contagion model. In this model, hate spreads like a virus through a social network. This model, however, fails to capture the fact that people do not acquire hatred from a single (...) infectious contact. Instead, it builds up in a person's beliefs and attitudes through time until the infection reaches a level where the subject themselves becomes a generator of hate speech. Second, to accommodate this, I consider an alternative model known as complex contagion. I argue that not only is a complex contagion model more explanatory and predictive, but it can be used to explain why certain features of social media cause it to be a promoter of hate. I conclude by sketching some mitigation strategies. (shrink)

The paper offers a critical discursive and pragmatic analysis of a corpus of hateful Facebook and Twitters status updates of politicians, political activists and voters in the 2016 pre-and-post election period, in Macedonia. Aiming to determine how power is exerted on social media, the paper focuses on identifying the stance social media users take when posting messages with political content. The analysis first attempted to unveil what speech acts the hateful posts are predominantly composed of, what roles the authors (...) of the posts normally assume, who the hateful political discourse in the given socio-political context is directed to, as well as what are some of the predominant linguistic strategies underlying the analysed hateful comments. The results show that, by using mostly assertive and expressive speech acts, social media users assume mainly the roles of analysts and judges and only subsequently the one of activists, they mostly address politicians directly and they use a lot of negative lexis, rhetorical figures and boosters as interpersonal metadiscourse markers to express their negative stance and exert power and dominance. (shrink)

Purpose The purpose of this paper is to present a quantitative and qualitative synthesis of the diverse academic proposals and initiatives for preventing and eliminating hate speech on the internet. Design/methodology/approach The foundation for this study is a systematic review of papers devoted to the analysis of hate speech. It has been conducted using the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) protocol and applied to an initial corpus of 436 academic texts. Having implemented the suitability, (...) screening and inclusion criteria, this corpus was refined to a sample of 74 articles. Findings The main subject categories studied in this corpus of academic research are legal issues and social media. In the majority of the articles, the use of hate speech via social media is associated with five typologies: religion, cyber racism, political slurs, misogyny and attacks on the LGTBI community. The absence of ethical reflection is one of the major shortcomings of IT-focused research and analysis devoted to online hate speech. Practical implications To date various systematic reviews have been presented, and they focus on detecting or describing hate speech. These have used either the search appraisal synthesis analysis framework or the Cochrane network. The PRISMA protocol was applied for this study, and both Scopus and texts in German were included. To date no major, rigorous systematic review has been undertaken of proposals to combat hate speech. Originality/value The link between hate speech and poverty has not been studied in depth within the academic sphere. Tolerance and ethical compassion are not granted the attention they merit when it comes to analysing the phenomenon of hate speech. (shrink)

This paper reports on a manual monitoring of online representations of LGBT persons in the Republic of Cyprus for the period April 2015–February 2016. The article contextualizes the prevalence of “hate speech” in online Greek Cypriot comments against LGBT individuals, and, more generally, against non-heterosexuals. Adopting a Foucauldian position vis-à-vis the social and discursive construction of sexuality, we outline, first, the socio-historical context with a focus on LGBT rights in the Republic of Cyprus and the nationalistic project (...) construing sexualities. We then examine the different levels of discursive discrimination practices, providing a snapshot of the types of “hate speech” referring to this topic typically found in such an environment. The focus is on identification of the frames used to construct LGBT identities, and their perception.We use in our title the word subject as defined by post-modernists and by Butler in particular : subject refers to “a socially produced ‘agent’ and ‘deliberator’ whose agency and thought is made possible by a language that precedes that ‘I’. In this sense the ‘I’ is produced through power ”. This paper focuses on the socially produced definition of the LGBT community in the context under study. We thus address the way in which sexuality is constructed within a compulsory and hegemonic heterosexuality and heteronormativity. We analyze our data i.e. comments focused on the LGBT community, with corpus linguistic tools as well as through a qualitative examination of the identified frames. Our analysis confirms an interface between nationalism and compulsory hegemonic heteronormativity in the Republic as well as the influence of the Orthodox Church and its beliefs. (shrink)

Patient To examine neural responses to aurally-presented sentences, a sparse imaging technique was used to minimize interference from scanner noise. The patient was played a single sentence (or noise-equivalent) in the 7.4s silent period before a single 1.6s scan with stimulus timing jittered relative to scan onset. There were 118 spoken sentences trials, 59 signal correlated noise trials, and an additional 60 silent trials for the purpose of monitoring data quality. The signal correlated noise stimuli had the same duration, spectral (...) profile and amplitude envelope as the original speech, but were entirely unintelligible (S1). The experiment was divided into three sessions of 79 trials with events pseudo-randomly ordered within each scanning session. The sentences were presented using a MRI compatible auditory stimulusdelivery system (Resonance Technology, Northridge, CA), with insert earplugs to further attenuate scanner noise. DMDX software running on a Windows XP PC was used to present the stimulus items. (shrink)

Data from academic seminars conducted through Skype textchat is analysed in this chapter, and the focus is on examples of how users mark community membership. Membership is marked explicitly by using pronominals and the metonymic use of the seminar group name. It is also marked implicitly by using reduced forms, which are stereotypical examples of a textchat speech style. I argue that these are markers of stance-taking, where community membership is recovered pragmatically as a weak implicature. Dis-alignment with the (...) community is also seen, as individuals can also implicate their independence as setters of linguistic norms. (shrink)

This article identifies five genres of anti-queer hate speech found in The Australian’s Facebook comments sections, exposing and analyzing the ways in which such comments are used to derogate cisgender and heterosexual women. One may be tempted to think of cis-het women as third-party victims of queerphobia; however, this article argues that these genres of anti-queer speech are, in fact, misogynistic. Specifically, it argues that these are instances of cis-hetero-misogynistic hate speech. Cis-hetero-misogyny functions as the “law enforcement (...) branch” of a cis-hetero-patriarchal gender order. Given the existence of such an order, it is clear that cis-het women’s liberation is inextricable from queer liberation. This article argues that to facilitate allyship and challenge this gender order—the order that elicits such hate speech acts—we need an epistemological revolution in the way we recognize and re-cognize human difference. (shrink)

Speech recognition technology has played an indispensable role in realizing human-computer intelligent interaction. However, most of the current Chinese speech recognition systems are provided online or offline models with low accuracy and poor performance. To improve the performance of offline Chinese speech recognition, we propose a hybrid acoustic model of deep convolutional neural network, long short-term memory, and deep neural network. This model utilizes DCNN to reduce frequency variation and adds a batch normalization layer after its (...) convolutional layer to ensure the stability of data distribution, and then use LSTM to effectively solve the gradient vanishing problem. Finally, the fully connected structure of DNN is utilized to efficiently map the input features into a separable space, which is helpful for data classification. Therefore, leveraging the strengths of DCNN, LSTM, and DNN by combining them into a unified architecture can effectively improve speech recognition performance. Our model was tested on the open Chinese speech database THCHS-30 released by the Center for Speech and Language Technology of Tsinghua University, and it was concluded that the DLD model with 3 layers of LSTM and 3 layers of DNN had the best performance, reaching 13.49% of words error rate. (shrink)

Online service providers —such as AOL, Facebook, Google, Microsoft, and Twitter—significantly shape the informational environment and influence users’ experiences and interactions within it. There is a general agreement on the centrality of OSPs in information societies, but little consensus about what principles should shape their moral responsibilities and practices. In this article, we analyse the main contributions to the debate on the moral responsibilities of OSPs. By endorsing the method of the levels of abstract, we first analyse the moral (...) responsibilities of OSPs in the web. These concern the management of online information, which includes information filtering, Internet censorship, the circulation of harmful content, and the implementation and fostering of human rights. We then consider the moral responsibilities ascribed to OSPs on the web and focus on the existing legal regulation of access to users’ data. The overall analysis provides an overview of the current state of the debate and highlights two main results. First, topics related to OSPs’ public role—especially their gatekeeping function, their corporate social responsibilities, and their role in implementing and fostering human rights—have acquired increasing relevance in the specialised literature. Second, there is a lack of an ethical framework that can define OSPs’ responsibilities, and provide the fundamental sharable principles necessary to guide OSPs’ conduct within the multicultural and international context in which they operate. This article contributes to the ethical framework necessary to deal with and by endorsing a LoA enabling the definition of the responsibilities of OSPs with respect to the well-being of the infosphere and of the entities inhabiting it. (shrink)

Modules, as Marr and Fodor conceive of them, lie between sensory and central processes. Modules have the functional property of representing that portion of the world which turns them on, and nine non-functional or structural properties that facilitate carrying out that function. Fodor has proposed that the processing of linguistic information is carried out by a language module , which therefore has the functional and structural features of modules. We argue that the proposed LM does not have the functional property (...) of modules in general . And we argue that Fodor's candidate for the output of the LM, interpreted syntactic form, does not satisfy important structural properties of modules . We propose another candidate, speech act potential, and argue that it fits almost all of Fodor's conditions . We next report on some pilot sentence completion studies suggesting that speech act information can influence the course of a parse and hence are a part of the LM . Finally, we outline possible experiments to test the modularity of speech act information by online methods of priming. (shrink)

In this post, Teresa Marques discusses her recent article in Journal of Applied Philosophy on whether hate is an essential component of hate speech. [blog post].

The impact of Online Social Networks on human lives is foreseen to be very large with unprecedented amount of data and users. OSN users share their ideas, photos, daily life events, feelings and news. Since OSNs’ security and privacy challenges are more potential than ever before, it is necessary to enhance the protection and filtering approaches of OSNs contents. This paper explores OSNs’ threats and challenges, and categorize them into: account-based, URL-based and content-based threats. In addition, we analyze the (...) existing protection methods and highlight their limitations and weaknesses. Based on that, we propose a comprehensive, user-level, proactive and real-time OSNs’ protection system, called Hybrid Real-time Social Networks Protector. HRSP has three components; a user-level security protocol and two classification models. The protocol defines a structure for OSN’s cryptographic services, including encryption, access control and users’ authentication. The classification models employ machine learning, black lists, white lists and users’ feedback, in order to classify URLs into: Benign, Risk and Inappropriate classes, and contents into: Benign, Hate speech and Inappropriate classes. We constructed two data sets of 150,000 URLs and 22,000 tweets to build and test the two classification models. Results show an overall accuracy of 93.2% for the URL model and 84.4% for the content model, while the protocol implementation produces compatible size and time overhead. The components of HRSP are integrated and have compatible design with OSN platforms. (shrink)