The Case for Children’s Linguistic Scaffolding

Clark and Hecht argue there are three mechanisms by which “er” is picked up by students: semantic transparency, singular-meaning affixes, and productivity. The test itself is intriguing, but a look into how students various reacted when they did not use “-er” was in some ways more revealing. Specifically, they resorted to compounds and suppletive terms, rather than simply ‘erroring out’: for example, ‘pull-things’ for someone who pulls things and ‘cleaner-people’ for someone who cleans people.

What this evidence suggests is that semantic transparency is the dominant effect of the three that Clark and Hecht identify: that is to say, without knowledge of singular-meaning devices that might produce the desired effect, or methods for making their language as efficient as possible, children resort primarily to trying to explain what they desire to convey as simply and cleanly as possible.

This is perhaps unsurprising: children seek to give meaning in as intuitive a way as they can. But it has fascinating implications for the way the mind’s ‘linguistic patterns’ are structured. Specifically, it does seem that absent a guiding linguistic mechanism, humans still display a disproportionate tendency towards very specific forms of expression. This is why agents were formed disproportionately with compounds and instruments disproportionately with suppletive terms. With no linguistic priors, it is not clear to me why one would do that: ‘picking-thing’ and ‘picking-man’ both seem reasonable compounds.

The suspicion that might therefore arise is that there is some sort of universal grammar or ‘linguistic sense’ making the determination for children that agents require compounds and instruments require suppletive. This also explains why ‘er’ takes time to become ingrained, as Clark hints at on p21: it might “hav[e] agentive meaning only” and therefore be difficult to universalize. 

What defense might we mount against those who argue this is an instance of instinctive linguistic knowledge? Perhaps it is the case that social norms and the behavior of parents and adults in conversation primes children to believe, for example, that instruments are specialized and therefore necessitate a specialized word. But to demonstrate this for certain, we would need more cross-cultural evidence. If it transpired that, regardless of a language’s cultural nuances, children displayed uncanny similarity in the ways they chose to structure language, it seems entirely plausible that we are given at least the scaffolding that is necessary to structure sentences – scaffolding which exposure to speech, or to ‘-er’ and similar affixes, then fleshes out.