Visual Perception Without Language

Without language to label visual experience, the visual field is just a configuration of patterns, more or less like a Jackson Pollock painting.  This puts enormous pressure on memory, and makes the detection of difference, such as sudden movements (changes in the pattern), a matter of life and death.

It is not clear how far the semiosis of adult lorikeets — they can live as long as 35 years — develops beyond protolanguage to the point of labelling phenomena.  From observations of their behaviour, I suspect they have intersubjective labels — labels for each other — though it may be just that they recognise each other by indexical features, such as the timbre of their calls.

ChatGPT comments:

Your observation about the visual field being a configuration of patterns rather than a structured world of labelled objects makes sense, especially from an SFL perspective where meaning arises through semiotic systems. Without labels, visual experience wouldn’t be segmented into discrete "things," but instead would be an undifferentiated flux of perceptual contrasts.

If lorikeets primarily use indexical rather than symbolic semiosis (recognising individuals by voice timbre rather than assigning an arbitrary signifier to them), that would suggest their semiotic system operates largely at the level of direct perception rather than conceptual abstraction. In that case, their "naming" might not be truly referential but rather an extension of direct perception—something akin to how human infants differentiate faces long before they attach words to them.

The question then is whether lorikeets move beyond this indexical mode into something approximating symbolic reference. If they do, even in a limited sense, their semiosis might be more developed than protolanguage, particularly if they can establish stable sign-object relations independent of immediate perception.