How else can thought be described if not through language? I don't know what you mean by "symbolic constructs." Symbols are the foundation of language—they're not the opposite. There is no sense in which symbols exist outside of at the very least a protolinguistic system. Once you begin to associate sensory data with meaning, you are doing the work of creating language. To analyze or describe cognition, we must use language, which organizes symbols into meaningful constructs. That thought occurs without language is not even wrong per se. It's unfalsifiable, which frankly is worse than being wrong in a scientific context. As Wittgenstein puts it, 'The limits of my language mean the limits of my world.' Without language, discussing thought is impossible, making the claim that thought occurs without language scientifically untenable. It is an attempt to position thought as the transcendental signified.
Yes, you need language to describe (discuss) something. But not everything that exists must have a description. Furthermore, meaningful does not require organization.
If you stand outside under the sun, do you have to be able to write the word "sun" in order to feel warm?
You're sidestepping the problem. Feeling warmth is a sensory issue. Connecting the fact that you're feeling warm with the fact that you're in the sun is cognition. In order to do that, you are doing the work of creating language. Sun equals warm.
