Is language change universal, ongoing, and arbitrary?

Question

Learning that arbitrariness from Saussure means there is no logical connection between the sound of morpheme and its meaning. But can we brain storm about this topic a little bit?

When it comes to the word "change", I'd rather put it in a larger way which is "development". I want to show an example: If you learn Chinese, then you would know actually mandarin is the one of "Chinese". If you go to the east and south of China, you would find, the so-called Chinese can't help you at all. Especially the sounds part, they are all different. The reason of this phenomenon is in ancient time, people of north of China moved to the different places, and settled down. The same (at vocal, shape, meaning) language changed through time.

In this case, the same original language contains dialects. These different dialects develop in their own ways among different social groups. Then why can't I agree the opinion as title saying? Language change is arbitrary. Hypothesis of mine: Language change is not absolute rule-governed. In some ways, it is arbitrary too.

Do studies suggest that language change is arbitrary, or are their rules or at the least generalisations that can be made?

Answer 1

I will first point out that SE is designed for people to ask specific questions that get specific answers which are either right or wrong. It is not designed to people to vaguely invite to chat and give opinions. Given that, I assume that you are asking how the idea of "arbitrariness of the sign" plays out in the domain of language change.

The idea relates to the connection between form and meaning in a synchronic grammar: that there is no principle of language that says that the word for dog is "dog" (proof: because sometimes it's beana, chien, Hund, mbwa or kəlβi). There is a mild challenge to this thesis, because there are sound-symbolic relations whereby a small thing that makes a high-pitched sound is more likely to be called a "titi" that a "dudu", and vice-versa with big low-pitched sounds. But I think these kind of supposed counterexamples are not so clearly counterexamples, they are examples of something else, namely the confounding factor of language change.

At the most basic level, the relationship between form and meaning over time is not arbitrary, instead, form(t)=form(t') and meaning(t)=meaning(t'). It is non-arbitrary that the current English word for dog is "dog", because it was "dog" 100 years ago; and that relationship between form and meaning extends back for many generations. At some point in the past (the details are far enough in the past that we're not sure how this happened), a word "dog" came to represent that meaning, and that replaced an earlier word along the lines of "hound" (which now has a more specialized meaning).

The field of historical linguistics has developed numerous tools for understanding why language changes, and I think the simplest explanation is that there are conflicting psychological "ideals" which cannot all be true, so social conventions about language develope to reduce random fluctuation. You can study this happening right now in the field of sociolinguistics, when you notice that people differ in how they talk in subtle ways, and (harder to study) they differ from themselves over time. The reasons are not arbitrary, but they are also not linguistic, they are social.