The orthodox neuroscientific approach to the study of consciousness is paradoxical: scientists examine consciousness and, infatuated with neurons, claim that brains must be the source of everything, and then declare consciousness epiphenomenal. Empirical expediency can provide spectacular scientific answers. Yet, one should make sure it is not to philosophically ill-posed questions. Investigating "how experience ensues from the brain", rather than exploring a factual claim, betrays a philosophical commitment. Here we examine the thesis of parallelism between cerebral and mental states: "consciousness (...) tells no more than what is going on in the brain; it only tells it in a different language". We retrieve an argument by Henri Bergson to expose the fundamental self-contradiction of parallelism: it forces the idealist to sustain that "the part is the whole", and the realist that "the part subsists when the remainder of the whole vanishes". We then recast Bergson's image-movement theory from Matter and Memory as a way forward to overcome the conceptual dead-end of parallelism. Consciousness is real. So is its special relation to the brain. Differentiating between solidarity and equivalence offers an alternative point of departure for an understanding of consciousness that does not, from the outset, outlay a false problem. (shrink)