Ellipsis & The Syntax of Silence

Generativism may the most widespread school of thought in linguistics today. From its beginning in 1957, every 7 years it has been transformed to adapt to scientic criticism, until its last transformation: minimalism, in 1990. It has captivated the minds of numerous linguistics departments, many universities from all over the world, and it receives support from many institutions.

So many are those who embrace the growing religion of the Galactic Empire AKA "Minimalist Program". I am not the only one who likes to imagine its leader, Noam Chomsky, as Darth Vader, fighting the resistance of other models of linguistic analysis, with its powerful army of prestigious universities.

I also like to imagine linguists like Martin Haspelmath embody the figure of Luke Skywalker, head of the resistance, fighting for scientific clarity against this Minimalism Empire that defends the existence of complex invisible structures of languages. You can join the resistance by following his blog, and know more about his fight in his post from 2018 "Is generative syntax simply a useful descriptive tool?".

On of the lastest important figures of the Minimalist Empire is general Jason Merchant, whose 1999 PhD dissertation "The Syntax of Silence" provided a extension of the model that allowed to analyse ellipsis in Minimalist terms. This theory has been widespread as a Death Star, obscuring other models of analysis... until the arrival of Peter W. Culicover and Ray Jackendoff's Simpler Syntax.

Anyway, it is interesting to critically approach minimalist analysis, and contrast them. In this blog from the Minimalist Galactic Empire, The Ling Space, the author presents a 10 minute overview of Merchant's analysis:

This video sketches the mechanics of Merchant's idea of ellipsis as a "silencing", or phonetic elision of a structure. It analyses ellipsis as a process where the remnant has been topicalised or fronted, and the rest of the structure has been elided because of its status as known information. 

In this way, Merchant's idea gives a satisfactory answer to the ellipsis phenomenon, because it allows the data to be analysed under the model itself, and this is the reason why the answer seem so satisfying: a mecanic explanation is provided that fits into an unified recognised frame of analysis that can account for every fact in language. The explanation seems them acceptable, but a critical linguist facing this theory would keep an attentive eye to remark the non-proven premises built in this theory: 

1. First, the fact that syntactic structure is linear: Phrase structure grammar models like the Minimalist Program presuppose that syntax has a fixed, universal order, and every difference from that order has to be interpreted as a further operation from this basic, primary word order. 
2. Second, the conditions in which elipsis can take place, the syntactic restrictions described as connectivity effects, are interpreted as the result of operations from that theoric primitive basic order.

Therefore, what data shows is that ellipsis has a number of constraints, but interpreting them as proof of the invisible mechanics of the invisible structure of a theoretical fixed order... doesn't seem more like a explanation of the model than an explanation of the data ?