I don't think only if one has seen it performed, although most powerfully to do so. We approach a script as a script, not only or even primarily as a written text, but as a blueprint for a performance, even if that performance is only the imagined one inside our head.
Even then it doesn't exist in absolute fullness because there are always an infinite number of potential different performances - and the richer and more complex the play, the less possible it is for any single performance to dig out all the possible interpretations contained within the script.
I said when we "consider it being performed"--not in any single performance, which would be as you say.
There might be irreconcible differences on a theory level. If one imagines Ros. wearing blue and the script says nothing, then it'd be idiosyncratic to consider that a feature of the text if you cannot share that interpretation with any interpretative community with a population greater than one (although online--you might be able to find one!)--which isn't to say the interpretation is wrong, per se. But when interpretative communities can agree on interpretations of the text, it is to those interpretations we refer to when we speak of the text, for the text itself is either squiggles on a page or pixels on a screen, in either case in need of interpretation to imbue them with meaning. (Cue Jubal Early here.)
no subject
Even then it doesn't exist in absolute fullness because there are always an infinite number of potential different performances - and the richer and more complex the play, the less possible it is for any single performance to dig out all the possible interpretations contained within the script.
I said when we "consider it being performed"--not in any single performance, which would be as you say.
There might be irreconcible differences on a theory level. If one imagines Ros. wearing blue and the script says nothing, then it'd be idiosyncratic to consider that a feature of the text if you cannot share that interpretation with any interpretative community with a population greater than one (although online--you might be able to find one!)--which isn't to say the interpretation is wrong, per se. But when interpretative communities can agree on interpretations of the text, it is to those interpretations we refer to when we speak of the text, for the text itself is either squiggles on a page or pixels on a screen, in either case in need of interpretation to imbue them with meaning. (Cue Jubal Early here.)