But we must not fail to lay special emphasis on one conclusion to which analyticwork along these chains of memory has unexpectedly led. We have learned that nohysterical symptom can arise a real experience alone, but that in every casethe memory of earlier experiences awakened in association to it plays a part incausing the symptom. lf - as I believe this proposition holds good withoutexception, it furthermore shows us the basis on which a psychological theory ofhysteria must be built.