This isn't the typical Pratt's lemma. In his notes$^\color{magenta}{\star}$, Pratt claimed the following without proof:
A simple case of interference is given by a Chu space having a constant row. If it also contains a constant column, then the two constants must be the same. Thus if [a Chu space] $A$ has a row of all 1’s it cannot also have a column of all 0’s. And if it has two or more different constant rows then it can have no constant columns at all.
The typesetting seems to suggest that this might have been proved as a "Proposition 1.1" or similar, but no proof is given. Could somebody locate or give a proof, please?
$\color{magenta}{\star}$ Vaughan Pratt, Chu spaces, Notes for the School on Category Theory and Applications, University of Coimbra, Portugal, July 13-17, 1999.
The typsetting issue you refer to is that $(i)$ all types of cited statement (Definitions, Theorems, Propositions, etc.) except examples are counted at once but $(ii)$ Definitions specifically don't get "leading numbers" indicating their section of origin. So the results in the paper go, in order: Definition 1, Proposition 1.2, Corollary 1.3, Corollary 1.4, Definition 5, etc.
So there's no missing proposition, and my read of the text is that the fact in the OP is just taken as self-evident.