In natural languages some syntactic structures are simpler than others. Syntactically complex structures require further computation that is not required by syntactically simple structures. In particular, canonical, basic word order represents the simplest sentence-structure. Natural languages have different canonical word orders, and they vary in the degree of word order freedom they allow. In the case of free word order, whether canonicalword order plays any role in processing is still unclear. In this paper, we present behavioral and electrophysiological evidence that simpler, canonicalword order preference is found even in a free word order language. Canonical and derived structures were compared in two self-paced reading and one ERPs experiment. Non-canonical sentences required further syntactic computation in Basque, they showed longer reading times and a modulation of anterior negativities and P600 components providing evidence that even in free word order, case-marking grammars, underlying canonical word order can play a relevant role in sentence processing. These findings could signal universal processing mechanisms because similar processing patterns are found in typologically very distant grammars. We also provide evidence from syntactically fully ambiguous sequences. Our results on ambiguity resolution showed that fully ambiguous sequences were processed as canonical sentences. Moreover, when fully ambiguous sequences were forced to complex interpretation by means of the world knowledge of the participants, a frontal negativity distinguished simple and complex ambiguous sequences. Thus the preference of simple structures is presumably a universal design property for language processing, despite differences on parametric variation of a given grammar.