It's often said that children seem to learn languages so much more quickly and effectively than adults. Yet adults do have some advantages of their own, as we'll see.
In the 1970s it was theorised that children's success was down to the notion that there is a critical period for language learning (pre-puberty). Once learners pass this period changes in the brain make it harder to learn new languages. Many took this critical period hypothesis to mean that we should get children to start learning other languages at an earlier stage. (The claim is still picked up today by decision-makers arguing for the teaching of languages in primary schools.)
Unfortunately, large amounts of research on the hypothesis has failed to confirm it. We cannot say for sure that younger learners are better (e.g. Ortega, 2009; Dörnyei, 2009). It's true that younger learners seem to cope more easily with pronunciation, whereas older learners usually maintain a "foreign accent" for many years. But other factors may explain why young children seem to learn a language with such apparent ease. In naturalistic settings young children are generally highly motivated to learn. They receive huge amounts of input at or above their level of knowledge as well as a vast amount of practice. Language learning at this stage is also rewarded by peer acceptance and the satisfaction of basic communicative needs. the same can't be said for many older learners of languages in classrooms. As Dörnyei (2009) points out, a large range of factors is involved in successful second language acquisition. Age may not be the most important one. What's more, there are recorded examples of unsuccessful child learning and successful adult learning.
Adults have the advantage in other respects. Older learners are good at vocabulary learning, for example and make use of different cognitive skills to young children. Adults can learn more analytically and reflectively. They can be more autonomous in their learning. They can learn and apply rules and patterns (and, to add, may have some very specific needs to motivate them, e.g. learning for a specific professional need).
How might these issues influence our classroom practice?
I tend to be a pragmatist on pedagogy and often find myself taking a middle road. My own experience in a secondary education setting over many years in England, suggested to me that we can take advantage of the possible advantages which both younger children and adults possess. Beginners aged around 11 or under can certainly be trained to pronounce very accurately and I think this should be one of a teacher's a priority. My experience with adult evening classes confirms to me that older adults have genuine difficulty getting pronunciation right.
Despite the research which suggests that learners that grammar is not "teachable", i.e. learners acquire grammatical patterns in their own, to some extent predictable order (e.g. Dulay, Burt and Krashen, 1982), and that they develop their own "interlanguage"* (Selinker, 1972), I have to say that in a non ESL context, where students hear and see very little of the language being taught outside the classroom, then I didn't feel that classes were immune to the order in which I taught grammatical structures. I am not convinced that secondary students acquire in the same order as young children. Having read a fair bit of literature over the years, my hunch is that the brain responds in complex ways to a whole range of input, some of it highly structured with rules given to help, some of it far more implicitly through general exposure.
The naturalistic, "comprehensible input" route reaps huge dividends over a number of years - but don't expect quick returns from all students in terms of spontaneous output. My own feeling remains that once students reach intermediate or advanced level, then immersion in meaningful input comes into its own. (I have read some teachers argue the opposite by the way.) Research does support the idea, however, that if you want students to be quite accurate, then you need to focus to some extent at least on the form of the language - for example by teaching pedagogical grammar rules or flooding input with lots of patterned examples and getting students to notice patterns or rules.
So beyond the age of 11 or so I would also argue for significant "focus on form", but not so much that it overly reduces the amount of patterned input at the right level. Students often want to know how the language works and there is enough evidence around in the scholarly literature that students can make their explicit knowledge of the language implicit. Put another way, they can learn a pattern and practise it to the extent that it eventually becomes available for spontaneous use. Declarative knowledge can, to some degree at least, become procedural.
Context has to be taken into account. What are the aims of the class? What exam are you preparing for? What is the culture of the school? What are your own strengths as a teacher? (Many teachers, for no fault of their own, lack the linguistic skill to deliver lots of accurate, well-pronounced comprehensible input.) What are your students like? What is their aptitude for second language learning?
In sum, young children and adults share a natural ability to acquire languages, but each group has unique advantages. In secondary schools we can exploit all the available advantages to help our students succeed.
* An interlanguage is a version of the new language where a learner preserves some features of their first language, and may also overgeneralize some L2 writing and speaking rules.
Dörnyei, Z. (2009). The Psychology of Second Language Acquisition, Oxford: OUP.
Dulay, Burt and Krashen, S. (1982) Language Two, New York: OUP.
Ortega, L. (2009). Understanding Second Language Acquisition, London: Hodder.
Richards, J. (2015) Key Issues in Language Teaching, Cambridge; CUP.
Selinker, L. (1972), Interlanguage. International Review of Applied Linguistics, 10, 209–231.