Recent studies have shown how social media platforms provide online spaces for people to express negativity and hate towards refugees, especially those who fled during the refugee crisis of 2015. With a theoretical framework of postcolonialism, orientalism, eurocentrism and a linguistic power perspective this study examines whether the same applies to Ukrainian refugees that fled during the Russian-Ukrainian war that started in February of 2022. The aim is to ascertain how both the groups of refugees that fled during the refugee crisis and the Russian-Ukrainian war were constructed by Swedish Facebook users.  Through critical discourse analysis 61 comments written on the Swedish newspaper Expressen's Facebook page were analyzed in this study. The results show how six out of seven discourses contributed to the construction of a negative image of refugees that fled during the refugee crisis. The most common discourse show how they constantly were being mistrusted and accused of lying as opposed to the Ukrainian refugees who were constructed in a more positive manner. There was a presence of eurocentrism and the “Us” and “Them” construction in the comments. Refugees that fled during the refugee crisis constantly were being distanced while Ukrainian refugees were constructed as a part of the European “Us”.

The Internet was initially seen as a metaphor for democracy itself. However, commercialisation, incorporation into existing hierarchies and patterns of daily life and state control and surveillance appear to have undermined these utopian dreams. The vast majority of online activity now takes place in a handful of commercially owned spaces, whose business model rests on the collection and monetisation of user data. However, the upsurge of political action in the Middle East and North Africa in 2010 and 2011, which many argued was facilitated by social media, raised the question of whether these commercial platforms that characterise the contemporary Internet might provide better venues for political speech than previous types of online spaces, particularly in authoritarian states. This thesis addresses the question of how the commercialisation of online spaces affects their ability to provide a venue for political speech in different political systems through a mixed-methods comparison of the U.S. and China. The findings of this thesis support the hypotheses drawn from existing literature: commercialisation is negative for political speech but it is less negative, even potentially positive, in authoritarian systems. However, this research uncovers a surprising explanation for this finding. The greater positivity of commercialisation for political speech in authoritarian systems seems to occur not despite the government but because of it. The Chinese state's active stance in monitoring, encouraging and crafting ideas about political speech has resisted its negative repositioning as a commercial product. In contrast, in the U.S., online political speech has been left to the market that sells back the dream of an online public sphere to users as part of its commercial model. There is still hope that the Internet can provide a venue for political speech but power, particularly over the construction of what it means to be a political speaker in modern society, needs to be taken back from the market.