Constructionist approaches to language recognize and emphasize the role of innate cognitive capacities in enabling, facilitating, and constraining language learning through exposure to and experience with usage in social interaction. However, such innate cognitive capacities, which include shared attention, pattern detection, inference of intention in context, sensitivity to frequency, etc. together with other learning constraints are domain-general and and non-linguistic in that they are not dedicated to grammatical structures (Tomosello 2003; Goldberg 2006).

The innate view of grammar has over the years been changing its mind about exactly what is innate. The latest incarnation of Generative Grammar (e.g. Hauser, Chomsky, and Fitch 2002) suggests that the only innate elements dedicated to language are (1) the ability to generate recursive structures; (2) a capacity that connects recursive structures to physical linguistic acts and general cognitive processes.

It may be helpful to think about the debate between Goldberg (2013) and Adger (2013) on this issue:

Adger, D. (2013). Constructions and Grammatical Explanation: Comments on Goldberg. Mind & Language, 28(4), 466-478.

Goldberg, A. (2013). Explanation and Constructions: Response to Adger. Mind & Language,28(4), 479-491.

Dear David, this seems to be a sensible proposition - that humans have an underlying 'structural' propensity for language but that this is realised according to social circumstances. However, it is difficult to get to the 'innate' structures because humans only develop actual language with and through others so the evidence for the big 'G' in the mind is indirect. Would you agree that this distinction (between 'G' and 'g') is perhaps similar to de Saussure's distinction between 'langue' and 'parole'? The former involves an abstraction from and idealisation of the latter. While this is very useful and makes 'Grammar' and dictionaries possible, it arguably depends on the creative messiness of language in use. Social constructivists would argue that language works from the outside (social world) inwards (individual mind) rather than the other (innate) way round. I think your idea is worth pursuing but empirically and conceptually challenging. Best wishes Peter

Thanks very much Peter. This is a great encouragement that it is a question worth pursuing. Dave

This is a very interesting question that I have asked myself many times. As a constructionist, I'm not supposed to admit that some components of human cognition associated with language are actually innate. I recently came to a sort of solution to this conundrum (something I feel comfortable with at least), though. As I see things, the over-reliance on the linguistic system has prevented a more holistic analysis of how language and thought are linked. In my view, concepts such as embodied agency, which I explore in recent papers, account for the existence of hybrid linguistic constructions that encode non-linguistic information and help humans conceptualize reality. Agency and intention are thus the driving force behind constructional selection and not simply statistical computations (whether they are innately given or learned). I invite you to read the papers below.

Article Constructions as Triads of Form, Function, and Agency: An Ag...

https://www.cambridge.org/core/journals/english-today/article/taming-english-modals/DBD3B215FE69EC19FAB4151152E3F34A

Article Exploring attachment patterns between multi-word verbs and a...

https://www.youtube.com/watch?v=OZhzaG5OHE4

Thankyou Sergio for your reply. It's interesting to note that there might be some common ground between the innate view and the totally learned view of language. So that one can say, well yes language is socially learned but this can only happen if it maps onto some innate conceptual patterns.

I will follow up on the references. Thanks once again. Dave

Constructionist approaches to language recognize and emphasize the role of innate cognitive capacities in enabling, facilitating, and constraining language learning through exposure to and experience with usage in social interaction. However, such innate cognitive capacities, which include shared attention, pattern detection, inference of intention in context, sensitivity to frequency, etc. together with other learning constraints are domain-general and and non-linguistic in that they are not dedicated to grammatical structures (Tomosello 2003; Goldberg 2006).

The innate view of grammar has over the years been changing its mind about exactly what is innate. The latest incarnation of Generative Grammar (e.g. Hauser, Chomsky, and Fitch 2002) suggests that the only innate elements dedicated to language are (1) the ability to generate recursive structures; (2) a capacity that connects recursive structures to physical linguistic acts and general cognitive processes.

It may be helpful to think about the debate between Goldberg (2013) and Adger (2013) on this issue:

Adger, D. (2013). Constructions and Grammatical Explanation: Comments on Goldberg. Mind & Language, 28(4), 466-478.

Goldberg, A. (2013). Explanation and Constructions: Response to Adger. Mind & Language,28(4), 479-491.

Thankyou Zhuo for this learned answer. It seems that the debate is unsettled and on-going. It's difficult to formulate empirical research to substantiate the nativist approach in order to differentiate general innate learning structures from those that might be related uniquely to language.

I have recently thought about the inner and outer of language where basic grammar structures may well be lodged in mind therefore psychological whereas words are cultural and evolve in the social world. That is why they are able to change in meaning. However basic grammar does not change in terms of agency (which however can be hidden by a passive voice) and intentionality and perhaps time. Best wishes Dave

Thank you all. Probably we have individually varying capacities to learn and construct differnt grammars of different languages and even different theories of grammar.