In the four-chain distribution of tokens (G-chains and H-chains), the number of chains combining b with a equals the number of chains not combining b with a.

The first two sequences may be called G-chains and the other two H-chains. Moreover, Mellor assumes that all tokens of \(A, B\) and \(C\) are distributed among the four chains so that the number of chains is exactly the same, namely one fourth of the sequences. Mellor then defines a causal relation between two singular events \(a\) and \(b\) in terms of a situation \(k\) which makes \(b\) more likely to occur given \(a\) than without \(a\), i.e., \(\rP(b\mid a) \gt \rP(b\mid {\sim}a)\). But we c