Is Adding the Suffix "-er" Innate or Learned?

The reading this week Learning to coin agent and instrument nouns by Clark & Hecht focused on how children acquire the conventional adult devices for coining new agent and instrument nouns in English. Specifically, the paper discussed how children learned how to use the suffix –er. The study showed that the youngest children were inconsistent with their use of –er, slightly older children were more consistent with their use of –er, and the eldest children used –er consistently for both agents and instruments. The paper proposed three principles that interact to account for this phenomenon. The idea the use of –er becomes more consistent over time recalls another idea that we studied previously and an idea that I have become enthralled by: is language truly innate? Carnie notes “children still acquire language in the face of complete lack of instruction.” One conversation that Clark & Hecht include in their paper is the following:

Child: What’s that called?

Mother: A typewriter.

Child: No, you’re the typewriter; that’s a typewrite.

True to Carnie’s sentiment, this child will undoubtedly come to terms with the idea that the physical object is called a "typewriter" rather than a “typewrite.” However, if children are learning how to fill gaps in their vocabulary by constructing new word forms to carry meaning for which conventional forms have not yet been learned or happen not to exist through their real-world experiences, linguistic structure, and social reasoning, shouldn’t we conclude that language is learned? To me, the idea that children can better understand the suffix –er over time means that at least some aspects of language are learned. It all seems like a problem of defining what it means for language to be innate. I would argue that even if children have the innate ability to reason through the suffix –er given context, this does not mean we can conclude that language is innate. Rather this is an argument that supports their innate ability to reason, not to use language. Thus, these findings by Clark & Hecht are evidence that language is learned rather than innate.