AFAIK, use of language shows that the being understands that the symbols used in language are abstract representations, not just a cause-effect type thing, and is capable of communicating to other beings using those symbols.
Merely saying “I want food” when hungry does not count as language. That’s just uttering a certain combination of syllables to produce a certain desired effect. Learning that “broken toy” means a toy can no longer be used, and then referring to spoiled meat as “broken food” DOES count as language: the being is expressing an idea in a novel (i.e. untrained) way. It has clearly associated “food” with meat, and “broken” with useless.
In short- language is defined by the association and expression of individual objects and actions with abstract representations of such. Many animals can communicate in a cause-and-effect type manner, but that cannot be considered language.
Language has also been one of the largest steps in human evolution: it’s what’s enabled us to communicate, coordinate and survive, and was the platform that has fueled the rest of our intellectual development. Animals have shown no conclusive evidence of communication - see Cecil’s article, as I mentioned above.