Ideological cleavages or East-West divide in the European Parliament? Explaining MEPs’ positions on EU's response to the invasion of Ukraine and annexation of Crimea

What determines MEPs’ positions on the foreign policy of the European Union? There can be a range of factors that can influence a foreign policy position of an individual legislator. However, in this paper we aim to explore the impact of two variables. First, we investigate whether party ideology of an MEP matters in how they talk about the EU foreign policy. And second, we explore whether there are any significant differences between the MEPs from West European member states and those who represent East European, i.e., post-communist, member states. For this purpose, we have selected two significant external shocks for the EU: Russian invasion of Ukraine in early 2022 and annexation of Crimea in 2014. We have hand-coded twelve purposively selected parliamentary debates, four of which are directly related to the annexation of Crimea in 2014 and eight to the invasion in 2022. To analyze the data, we will use mixed methods approach and conduct our analysis in two steps. First, to evaluate whether or not the two variables matter for MEPs’ foreign policy positioning we will use multinomial logistic regression analysis. We code party ideology depending on to which party group in the European Parliament an MEP belongs. The East-West divide is coded according to which member state an MEP represents at the European Parliament. If the member state is a former communist country, then the MEP is assigned a code of Eastern Europe. As for the dependent variable, we have coded each sentence or a quasi-sentence of all interventions by the MEPs in twelve debates into one of six larger categories of EU foreign policy roles. These six categories include the following: geopolitical power, civilian power, normative power, ethical power, transformative power, and inert power. While the first five categories are theoretically derived from scholarly literature, the last one, inert power, is an empirical category created during the coding procedure. Consequently, the dependent variable is categorical and has six values. Second, if we observe that one or both of our independent variables are statistically significant, we will qualitatively explore what differences they result in when MEPs talk about EU’s foreign policy role. If we find no effect of the two independent variables, we will try to turn to the literature to generate potential hypotheses about why this may be the case. We may also observe that there is statistically significant effect in only one of the two cases, in which scenario we will again try to explore both avenues of interpretation, i.e., identifying qualitative differences and proposing potential hypotheses about why we observe this particular result.