We study the problem how to aggregate different opinions of   m decision makers who evaluate n objects with respect to a criterion into a single opinion of the group. In particular, each of the m decision makers evaluates the n objects pairwise with respect to the given criterion. In order to unify various approaches, we assume that the decision makers use the elements of an Abelian linearly ordered group (alo-group) to assess the relative importance of the two items in each pair of the n objects. Moreover, a decision maker (DM) can use a fuzzy subset of the alo-group to assess the relative importance whenever the DM is uncertain about the exact value of the assessment. Thus, the task is to compute a joint priority vector of  m given n×n reciprocal pairwise comparison matrices with fuzzy elements, i.e. fuzzy subsets of an alo-group. In this paper, we consider several desirable properties of the priority vector – consistency, intensity, and coherence – and we propose a new algorithm to compute priority vectors satisfying these desirable properties.