Today I’m a bit sick of the assumption “proper” scientists make that linguistics is not a “proper” science.
Science (from Latin: scientia meaning “knowledge”) is a systematic enterprise that builds and organizes knowledge in the form of testable explanations and predictions about the universe.
Let’s start with systematic. The various branches of linguistics form a systematic investigation of language. Syntax investigates how sentences are structured, phonetics and phonology the shape of the mouth when producing sounds, morphology the investigation of how words are structured and new words are created. Semantics and pragmatics seem to be a bit more fuzzy, admittedly.
The key here is “testable explanations and predictions”. Two examples from a post of mine elsewhere:
Morphology – the study of how words are built up. Predictive in that the rules of word building have been laid down by study enough to predict how a newly coined word will give rise to new derivational forms. Eg. Laser (acronym) was predictably backformed into the verd “to lase” because we know that verb forms with the derivational suffix -er denote the agent of the verb. Same as we got “to peddle” for pedler, “to beg” from beggar we get “to lase” from laser.
Phonetics – We know that speech is continuous, not discrete like the alphabet would have us believe. Try saying “inpressive” rather than “impressive”. It’s difficult because at the end of a sound our lips and tongue are already ready for the next sound. With the “p” being a bilabial plosive and “n” being alveolar nasal, the lips are in the wrong position to produce it, therefore we automatically substitute the bilabial nasal “m”. This means that we can predict that, no matter the language and therefore alphabet, “n” in front of “p” has to sound like “m”. I can illustrate this using Japanese. The only non open syllable sound in Japanese is ん which sounds like the n on the end of the word on. If it is front of a “p” syllable, for example in the word さんぽ (sanpo) meaning “a walk” it comes out sounding like sampo, regardless of the fact that ん represents n.
So at least two branches of lingusitics are systematic, and make falsifiable predictions. Therefore linguistics is science.