Some thoughts on what is LANGUAGE SCIENCE, literally.
0.0 Preface of the Preface
Many years later, as he faced the grad admission committee, Yixuan Wang was to remember that distant afternoon when the undergrad admission staff from Peking University asked him why he chose to pursue a linguistics major. The cold-looking staff was actually the head of the undergrad admission office and the dean of the School of Psychology, as I later discovered.
Well, I said 'I just like it'. Lucky me.
And I may still be fond of linguistics now, after three years of terrible and disappointing mal-education. But, why linguistics? I still don't have an exact answer, but at least I get a slightly clearer picture of linguistics. So, I think it is kind of necessary for me to explain what I pursue when talking about linguistics, before seeking the answer to why.
Obviously, the name of the series is taken from the famous Chomsky quote(1957), but with some meanings. I am trying to be 'colorless', presenting objective perspectives, but I am still too naive, or, 'green', to fulfill that goal. So these ideas are only mine, limited by my current view.
0.1 Language Science
My basic stance is clear, and I don't expect I will ever change my mind.
Linguistics is the science of human languages. If it IS not, then it SHOULD HAVE BEEN, and it SHALL BE.
I am not familiar with the philosophy of science, but I am pretty sure we need science for understanding the world in a relatively objective way. Languages can be artistic, but to understand what powers them, we have to inspect them scientifically.
This means, all the methodology and techniques used in other science, apply to linguistics - or I would prefer to call it, language science. Let's assume the term 'linguistics' is just a short form of 'language science'.
We are not going to define language in this post because it seems trickier than a mere statement. Let's assume that language is some sort of intermediate representation that encodes and powers our thinking AND communication with 'the other'. Maybe we'll discuss that in some future posts (if any).
0.2 Subject of Language Science
Language science concerns languages. That's almost nonsense. But which utterances (or utterances not spoken out) qualify for the word 'language'?
In structuralism, only those materials recorded are considered legal projections (parole) of an abstract language (langue) shared by the whole community. Only the abstract language behind these perceived behaviors is the authentic subject of language science.
In Chomskyan generativism, utterances are considered merely forms that can be produced (generated, as the name suggests) by a presumed innate physical facility in the human brain, under the guidance of a formal system. These systems can produce forms that are not yet uttered, but there are forms they are not capable to produce. So, the main focus should have been describing the formal system and the innate facility possibly powering it.
The stark truth is that human languages are way too complex and flexible. People coin new forms, even new systems, from time to time. Simplified models cannot describe them. We are NOT arguing that only theories which can explain everything are scientific, but denouncing facts that cannot be modeled by a particular theory is JUST NOT THE RIGHT WAY of doing science.
Although I have no idea how to define 'language' itself, linguistics needs to study (or to be able to study) more than we can now.
0.3 Top-down Approach
Surface forms of languages are always breathtakingly beautiful. That's why enthusiastic polyglots and passionate language nerds (including me, of course) exist.
But, they are superficial. The serious science of languages should dive into the mechanisms lying behind them. When playing with language forms (most likely phonology, morphology, syntax, and semantics), overfit can (or, always) happen. And we cannot guarantee or expect that one day, we can disentangle the full process from underlying mechanics all the way up to the surface forms we observed.
We, however, should still observe these superficial patterns (with extra caution in mind), as these output forms are the only concrete evidence in our hands.
0.4 Antireductionism
It is not a great idea to slip into your armchair and wait for the science of brains to develop and bust the science of languages one day, that's pure (theoretical) reductionism. I was once distressed by the gap between current theories of languages and real-world language phenomena, hoping that neuroscientists will explain language mechanics as ground truth and kindly suggest linguists apply for unemployment relief in advance.
Well, no. Although language is indeed part of human behavior and cognitive process, which is concerned by ethology and neuroscience, language is way beyond. To put it more clearly, it is not only an intrinsic system for reasoning and thinking but also an extrinsic system for interaction and communication. The latter part is the major breach in attempts to reduce linguistics to some sort of biology.
My physicist friends often preached the famous antireductionism point brought forward by Philip W. Anderson, which I think is quite fair (though I have no idea of symmetry breaking):
More is different.
0.5 mem::forget(linguists)
Since the very beginning of pretty unscientific language studies (which sometimes we still dub linguistics), researchers in this field borrow ideas from foreign fields ranging from relevant ones like mathematics (logics) and biology (taxonomy trees and evolution ideas) to exotic ones like chemistry (that valency stuff in argument structures) and so on. Although some ideas proved to be ridiculous, in general speaking, borrowing ideas and analogy are nothing unhealthy. Chomksy drastically revised his ideas on generativism several times, not because of getting things wrong time after time but being open-minded to incorporate new thoughts for the sake of his ultimate goal - revealing the cognitive mechanism and mechanics behind the human language.
Sciences are not xenophobic and ultra-conservative, cults are. Sciences are open to tweaks, fixes, refurbishments, and resurrections (let alone criticisms), but cults are not. Recently, linguistics often gets itself into trouble with artificial intelligence and data science stuff. That's a long story, but to put it short, there are clearly some lessons for linguists to be learned from modern-era alchemy.
The title of this section is a standard library function for leaking memories safely in the Rust programming language, which I think is a perfect metaphor for the preferred mindset working on such a rather nascent and tentative field. Personally, I don't stick to a belief (not practice) in any specific theory or descriptive framework. I'm fine to overthrow my own ideas or learn new things from other fields, as long as I am persuaded.
0.-1 Closing Words
I think that's all for this post filled with vacuous talks now, but that's only a small fraction of my green ideas. I chose the index of 0 to indicate this post is considered as (current) guidelines and basics of my academic views and purposes. Hopefully, we can dive into details in some later posts.