Many musicians performing an ensemble work would aim to achieve good, if not perfect, synchronisation of timing for the sake of an effective rendition (McCaleb, 2014). In this respect, Pau Casals’ and Mieczysław Horszowski’s 1935 recording of the second movement of Johannes Brahms’s F major Cello Sonata, Op. 99, is certainly intriguing, as asynchronies between the two instruments are clearly perceptible throughout extensive passages. Questions concerning the rationale behind these temporal discrepancies arise: To what extent were Casals and Horszowski intending not to play together? In what ways can we understand these asynchronies? What are their musical effects?

Analysis of timing issues that focused on the magnitude of the asynchronies as well as on tempo curves and durational patterns employed by the performers in this recording has shed light on the effects, purpose, and meaning of those asynchronies. This article illustrates that the temporal discrepancies between the cello and piano parts are neither the result of chance nor only local expressive devices; rather, they are critical elements in a highly distinctive construction of the Adagio Affettuoso from Brahms’s Op.99. They attest to an ongoing structural dialogue between the musicians that shapes and paces the piece in unexpected ways, prompting a reconsideration of ensemble playing as affording different, even conflicting attitudes. Furthermore, their potentially structural import leads to a critical broadening of the methods and parameters that music analysis has traditionally acknowledged. On these bases, a fruitful dialogue between music analysis and performance is put forward.