I asked this question on the homotopy theory chat, but I haven't got any answer - thus I decided to post it as a question here.
The question is rather historical. Let $X$ be a connected topological space. I know that nowadays we can prove that the James construction $J(X)$ is a homotopy model for $\Omega\Sigma X$ using the Bott-Samelson theorem (as, for example, in Neisendorfer's book "Algebraic methods in unstable homotopy theory"). Bott-Samelson theorem states that if $R$ is a PID and $X$ is such that $H_\ast(X;R)$ is torsion-free, then $H_\ast(\Omega\Sigma X;R)\cong T(\tilde{H}(X;R))$, where $T(M)$ denoted the free tensor algebra on the $R$-module $M$.
However, in his paper "Reduced Product Spaces" James didn's cite Bott-Samelson work, which was published earlier. So my question is - what is the historical relation between these two concepts? The connection was seen since the beginning, or was it found later?
