Some philosophers argue that something relatively complex must have predated language-use in humans otherwise language wouldn't have arisen in the first place. In other words, language couldn't have just come from nowhere. Some kinds of conscious activity (or even cognition) must surely have predated language-use. The question is: Which kinds of mental activity or cognition?
I suspect that some people would question the word cognition (i.e., if there were no language) for two reasons. One, language is used in most examples of cognition. Two, only a language-user would have risen to the level of cognition in the first place.
There are acts of consciousness (or mental events) which, arguably, don't involve language in adult humans. Mental imagery is one example.
For example, you can imagine a round blue shape being stuck on the surface of a black square. You can even imagine adding an extra nose to someone you know. Despite that, some philosophers may argue that such cognition (even if non-linguistic) would only be carried out by language-users. They may say that language even "infects" such language-innocent examples as the mental imagery cases just mentioned. After all, without the mastery of language, you may not even have a concept of a square and of a round shape. The same goes for adding a nose to a face. The mental imagery itself doesn't include the use of language; though perhaps only a language-user could think about such mental images and juxtapose them in the ways he does. (Mathematical equations being carried out in the head don't require words. However, would someone have reached that level without being a language-user in the first place?)
I'm not sure that philosophers (on their own) can answer questions as to what humans thought about (or how they thought) before they were language-users. (Weren't humans always language-users in some very basic sense?)
I mentioned the evolutionary fact (or possibility) that something must have predated language all those hundreds of thousands of years ago. Though we don't need to go that far back. Philosophers have also argued that here and now, before we say something (or even think something to ourselves), something must come before the articulation (even if a sub-vocalisation) or the use of a natural language. And, just as with our far-off ancestors, something must come before the actual expression in language.. or must it?
Some philosophers talk about an abstract proposition coming before its expression in a natural language. That may be true. Though that doesn't impinge on what goes on in an individual's mind and brain before he expresses himself in a natural language. Even if abstract propositions exist, this is a different question about what predates the expression in the mind-brain of the language-user (regardless of a realm of abstract propositions, if such a thing exists at all). One answer is the "language of thought" (LOT) theory first advanced by Jerry Fodor. Others provide a more naturalistic (or neuroscientific) account of what happens before we say or think something in a natural language.