The final dimension of this passage concerns leader-follower dynamics. It was one thing to know explicitly that we needed to try and ‘keep the momentum’ as a group between b.179-187, and to avoid overly predictable units of phrasing. It was another to do this while also capturing the sense that the first violinist was “dragging the sound forward, rather than constantly chasing it”. This metaphor also hints at the frustration one experiences if the cart gets ahead of the horse. An open disposition was once again key to unlocking these internal dynamics: to have ‘space’ to push or pull other voices requires them to provide it.

Staying with b.179-198, we found it helpful to unify our physicality in the bow, and to find a shared character of contact that would underpin our individual variation. This approach to the archaeology of recordings lies a long way from painstakingly identifying the timing of notes. This more ‘already whole’, embodied perspective on their playing offers an alternative ‘way in’ to the Czech Quartet’s sometimes peculiar trajectories. As in #1, we met with more success in ‘feeling’ their rhythm when our contact was a little firmer than one might assume from listening alone. But as before, they guard against the tendency for this greater contact to yield rhythmic stolidity.

In b.179-198 Florian violist Anna perceptively traced some of the stylistic contrast to the preparation and completion of phrases. From a performer’s perspective, the idea of preparation – especially when it involves the breath – is a usefully embodied explanation of the way in which ‘structural’ phrasing generally emphasises separation between perceptually salient units. This is paradigmatically on the level of the phrase, but it can work at either a smaller or larger scale.* She noticed the Czech Quartet’s uncanny ability to elide gestures of completion into gestures of preparation: in the passage above, for example, they ‘prepare’ the phrase ahead while they are still in the process of completing the preceding material.

In #22 we saw how treating asynchrony as ‘allowable’ in principle also changes the character of playing that is closely synchronised. Two examples following in quick succession – between b.155-171; and b.179-198 – build on this point. The first is easier to grasp, because its unfolding tensions develop progressively. First violinist Hoffmann presents new (hymn-like) material (b.155) in a flowing tempo that seems to ‘come out of’ the preceding music. That tempo is quickly undermined by next entry, however: it is as though second violinist Suk starts ‘holding’ his colleague, challenging him to resist the easy ‘flow’ he has already set up. The two-way valve concept is essential here, because at no point are these voices perceptibly ‘asynchronous’; yet in terms of their attitude towards the material they pull in different directions.

A particularly difficult transition for us to grasp was between b.47-52, when hard-edged martial gestures give way to delicate playfulness. This is a good example of the ostensibly less ‘structural’ outlook characteristic of the Czech Quartet’s generation, and of the idea that playing across phrase boundaries was an available option, if not exactly a norm. In this quicker movement we often found ourselves drawing attention to such moments of transition – surely because the Czechs’ handling of them seems so unusual and marked, from our perspective – and this resulted in a sense of conscious manipulation that was far more subtle in the original.

In #31, #32 and #33 we encountered accompanying bowstrokes that seem designed to invoke instability in timing and tone (e.g. b.25-29). The embrace of individual indeterminacy is significant for the copying process more generally, in that ‘planning to come apart here’ is a different kind of ‘strategy’ to that which likely generated the original performance. In the intervening time we generated many more insights into the potential modes of interaction between players.

This section offers a small taste of our evaluations in the passage b.39-59. We felt this take was getting closer, but the cello needed to be more ‘covered’; the inégale was too subtle – to the extent of being barely audible on listening back; vibrato was too present in second violin; and the violist remarked that far from hearing her own intentions back – which had seemed vivid at the time – the sound played back was ‘plain’, as if lacking ‘content‘ (especially by comparison with the Czech Quartet).

Their manner of ‘recovering’ time often felt unfamiliar and unpredictable, probably because it was never formulaic. We felt was that the concept of recovery must have been important to them on some level, but it was remarkably difficult to generalise about their strategies. This difficulty encouraged us to look again at how we were conceptualising timing and beat structure. Theorising about the idea of recovery sometimes gives the impression of a functionally independent (and regular) timing framework lying ‘above’ the performance, where the process of ‘paying back’ time is underpinned by a metaphorical pendulum that swings between adherence and deviance. Though appealing in theory, we found that this model did not accurately reflect the musician’s relationship to time in each moment of performance.

Admitting unevenness into the very foundations of one’s rhythm brought an unexpected benefit, in that it opened up expressive potential in the other direction. Playing very evenly, rather than being the de facto norm, could itself be much more ‘marked’, by comparison with more irregular surroundings. It is probably misguided to speculate about how the original players experienced the ‘affect’ of this, given the extent to which it was dependent on context – even for us. In b.47-48 of the Dumka, second violinist Josef Suk plays his undulating semiquaver figures considerably more evenly than in the surrounding bars, which are characterised by much greater tension – not just harmonically, but rhythmically, as a result of his much less predictable uneven inflections.

Hoffmann frequently breaks notated slurs in this passage and many others. It is impossible to tell if this is a relevant convention – in the sense that the meaning of such slurs was broadly different, and more fluid than later in the twentieth century – or if he just ran out of bow in the moment of performance, and preferred to prioritise the sound’s continuity by taking another.