In recent years, technology-scepticist views caution that internet technologies are far from utopian places or idealisations of Habermasian public spheres. There are growing concerns about the proliferation of harmful content online, including online hate speech, harassment, abuse, etc. In this thesis, I focus on one of the most concerning and harmful behaviours on Twitter and in politics more broadly: political incivility and intolerance, through a case study of political tweets about abortion in Ireland (2018) and the United States (2020). The thesis aims to enhance our understanding of the nature and dynamics of political incivility and intolerance in abortion discourse and on Twitter. I make three main contributions to research on these topics. First, I make a critical-normative contribution by discussing online incivility and intolerance in connection with deliberative and critical theorists, including Jürgen Habermas' theories of public spheres, deliberation, and civility as well as Rainer Forst's theories of pluralism and toleration. My thesis establishes a concrete normative vision of what democratic communications ought to be. Furthermore, this normative discussion compensates for the limitations of computer science-driven content regulations, which lack theoretical underpinnings in terms of what they mean by 'toxic,' 'harmful,' or 'uncivil' language in their models. Following Habermas' and Forst's theories, I argue that incivility can have some limited roles in deliberation whereas intolerance is incompatible with fundamental principles of deliberative and pluralist democracy. Secondly, I make an empirical-descriptive contribution to research, exploring the Irish and U.S. Twitter data both quantitatively and qualitatively. To study the topic from multiple perspectives, I employ an innovative methodological triangulation, combining computational text mining methods and manual qualitative text analysis. Quantitative big data analysis produces a general linear model to predict the relationships between incivility, intolerance, and diverse demographic, political, and communicative contexts, e.g. high-profile political events and issues, abortion issue position, issue partisanship, gender, anonymity, tweeting context. The qualitative analysis explores rhetorical patterns and types of political incivility and intolerance, unfolding how Twitter users employ political incivility and intolerance, and what assumptions and ideologies are embedded in them. The qualitative analysis reveals that Twitter users construct antagonism and false polarisation between 'Us' versus 'Them,' sabotaging reasonable deliberation and policy compromises. The cross-country comparative analyses indicate that political incivility and intolerance are cross-cultural concepts but also unfold in culture-particular ways with specific talking points and vocabulary. Thirdly, I discuss the prescriptive implications of my normative and empirical discussions and empirical findings for the health of deliberative politics, illuminating how we should understand incivility and intolerance and deal with them. I make five noteworthy observations for extensions and future scholarship: (1) there is a strong predictive relationship between political incivility and intolerance; (2) Twitter structures can hinder productive sublimation of anger to persuasive arguments; (3) a small set of hyper-active users dominate the uncivil and intolerant communications; (4) there is little relationship between anonymity, political incivility and intolerance; and (5) there is a link between political incivility, intolerance, and the rise of populism and reactionary backlash. I also make social impact recommendations concerning platform redesigns and civic education about online ethics. Through these three original contributions, this PhD thesis not only benefits ongoing scholarly knowledge and debates on digital culture and online user behaviours, but also contributes to lasting social impacts such as paving the way for future projects on social media platform redesign, the development of ethical and principled content regulation algorithmic models, and civic ethics education for members of public who use social media for political and activist purposes.