Previous research has shown that legislators defecting from the party line often take the opportunity to provide an explanation for their behaviour on the parliamentary floor. However, beyond anecdotal evidence, researchers have not yet looked into the justifications that legislators provide for their voting decisions. Do legislators provide well-grounded rationalizations of their defection or do they simply deliver empty statements to create a positive record for themselves? I address this question by studying a multilingual corpus of 151,460 oral and written voting explanations from the eighth European Parliament (2014-2019). Using natural language processing and statistical learning techniques, I measure the degree of reasoning in these voting explanations. The preliminary findings suggest that the extent of reasoning depends on the type of defection and the issue salience of the vote. The findings have implications for our understanding of the use of reason in political discourse and legislators' relationships with their party principals.