I and my colleagues have been extracting data from Russian media and social media to understand how both propagandists and social media users discuss the war. The first report is out. Two more will be published later. Some preliminary observations (15):
1/ The number of stories about the war on TV has been decreasing since Feb. Still a lot of propaganda, but twice less than in Feb-Mar. The key justifications for the invasion - ‘demilitarization’, ‘denazification’, NATO expansion, Donbas people - are discussed less and less. 
2/ ‘Demilitarisation’ and ‘denazification’ used by Putin to legitimise the invasion were frequently used on TV in the beginning of the war to be abandoned by mid-March.
3/ This is not something new. As others report, people were struggling to comprehend the meaning of these words, let alone to embrace them. Some officials can still use them out of inertia, but in general propagandists have failed to explain them and abandoned them.
4/ Other justifications - Donbas people and NATO - are more persistent. Protection of the Donbas people was and is used more frequently than others justifications. The number of references to NATO is lower compared to Feb as well except for the NATO summit in June.
5/ The use of the word ‘war’ is a subject to criminal liability. However, TV propagandists routinely use it! Just not in the way we expect them to.
6/ The war in Ukraine - the one fought by Russia with tanks and missiles - is mentioned rarely. Instead, TV channels discuss other ‘wars’ waged by the West against Russia - ‘information war’, ‘visa war’, ‘sanctions war’.
7/ The same with ‘crisis’. Considering deteriorating economy, you would expect them to pay some attention to economic issues. They do. But not to Russia’s economic crisis! They are discussing ‘food crisis’, ‘gas crisis’ - the challenges faced by other countries, not Russia.
8/ We call these crises and wars ’spoiler crises’ and ‘spoiler wars’. Russian TV channels endow words that seem undesirable with new meanings appropriating the vocabulary of critics.
9/ How about social media users? Both TV and social media users are equally interested in Ukraine. However, as noted by many before, the discussion is much freer on social media despite criminal liability for the word ‘war’.
10/ Despite the attempts to portray the ongoing events as a limited operation, social media users discuss it as a war. Judging by manually verified messages, this is also typical for supporters of the authorities.
11/ Users are also not very receptive to the official justifications. ‘Demilitarisation’, ‘denazification’, etc are less frequent. Except for the spike in July when a high-level official made a highly quoted statement in response to threats to destroy the Crimean bridge.
12/ Social media users are also less receptive to the rhetoric of ‘spoiler crises’ and discuss economic consequences less.
13/ However, they are quite receptive to polarising hate speech - especially the term ‘fake’. Since social media is a space where verification of information is a challenge, the language of ‘fakes’ spread by propaganda finds fertile soil in social media discussions.
14/ We used Scan Interfax for media (16k messages) and Brand Analystics for social media (408k messages). BA does not scrap data post factum, so no social media data for Feb - Jun. SI can’t scrap data from blocked media, so mass media corpus is skewed towards pro-regime media.
15/ Social media corpus includes only users who indicated Russia as a location. It helps to exclude Russian-speaking users from other countries, but heavily skews the corpus towards Russian domestic social media (Vk, Odn, etc)
• • •
Missing some Tweet in this thread? You can try to
force a refresh
