Let $\mathcal{L}$ be a first-order language, and let $\mathrm{LJ}^{-}_{\mathrm{list}}(\mathcal{L})$ denote the single-conclusion sequent calculus whose sequents have the form $\Gamma \Rightarrow S$, where $\Gamma$ is a finite list of $\mathcal{L}$-formulas and $S$ is either empty or one $\mathcal{L}$-formula. Assume that identity is the contextual initial rule $\Gamma,A,\Delta \Rightarrow A$, that the primitive logical rules are the usual intuitionistic left and right rules for $\top$, $\bot$, $\land$, $\lor$, $\to$, $\forall$, and $\exists$ with explicit finite list side contexts, and that left weakening, left exchange, and left contraction are not primitive inference rules. Assume also that these logical rule schemata are stable under uniform list-context reindexing: if one inserts the same extra formula occurrence into every antecedent list in a rule instance, swaps the same adjacent antecedent occurrences in every antecedent list in a rule instance, or identifies the same two adjacent equal formula occurrences in every antecedent list in a rule instance, the result is again an allowed instance of the same logical rule whenever the principal formulas and eigenvariable side conditions are unchanged. Then the following left structural rules are admissible in $\mathrm{LJ}^{-}_{\mathrm{list}}(\mathcal{L})$: insertion of an extra antecedent formula, swapping two adjacent antecedent formulas, and deleting one copy from two adjacent equal antecedent formulas.