Chirality is transduced, not determined

Every organism uses L-amino acids, and the uniformity is complete. Recent work has substantially accounted for how a local excess arises, how it survives racemisation, and how it spreads across the components of a mixture: a racemic fluid separates into homochiral domains held by an asymmetry in activation energy, with symmetry broken by percolation in a confined volume, and a single non-racemic component in a competitive ligation enantioenriches all its partners and products. In each case, however, the sign is fixed by a contingency to which the mechanism is indifferent, so what the chemistry delivers is not a homochiral world but a population of locally homochiral compartments with independently assigned signs. This paper argues that the bias RNA exerts on the handedness of the amino acids it carries is real but is not a determination. It is suppressible by simple manipulation, inverts with a mirror sugar, changes sign between ribozyme motifs, is essentially absent from non-enzymatic aminoacylation, and where bond-forming chemistry expresses a stereochemical preference at all, that preference favours heterochiral coupling. The sign was instead made universal at the merger of a peptide-bond generator with a population of specifically charged carriers, the same event proposed to have fixed the genetic code. The generator is indifferent to handedness and transmits what it is given; handedness is therefore transduced from carrier to peptide across a step with no preference of its own, converting a handedness maintained by ongoing chemistry into one inherited through copied sequence. A test is proposed: a published peptidyl-transferase-derived scaffold, charged with D- and with L-minihelices, should yield product whose handedness tracks the carriers'.